Lithos 119 (2010) 1–9

Contents lists available at ScienceDirect

Lithos

j ourna l homepage: www.e lsev ie r.com/ locate / l i thos

Paleokarst on the top of the Maokou Formation: Further evidence for domal crustaluplift prior to the Emeishan flood volcanism

Bin He ⁎, Yi-Gang Xu, Jun-Peng Guan, Yu-Ting ZhongKey Laboratory of Isotope Geochronology and Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Science, Guangzhou, 510640, China

⁎ Corresponding author. Tel.: +86 020 85290133; faxE-mail address: hebin@gig.ac.cn (B. He).

0024-4937/$ – see front matter © 2010 Elsevier B.V. Aldoi:10.1016/j.lithos.2010.07.019

a b s t r a c t

a r t i c l e i n f o

Article history:Received 8 October 2009Accepted 21 July 2010Available online 11 August 2010

Keywords:PaleokarstMaokou FormationCrustal upliftEmeishan large igneous provinceMantle plume

The ~260 Ma Emeishan Large Igneous Province (ELIP) in southwest China has previously been demonstratedto provide compelling evidence for pre-volcanic crustal doming in support of the mantle plume hypothesis.However this has been questioned by Ukstins-Peate and Bryan (2008) by showing hydrothermal magmaticactivity at the Daqiao section. To solve this argument, a detailed characterization of the contact between theEmeishan basalts and the Maokou Formation was carried out. The contact is shown to be an unconformity,which is characterized by paleokarst on top of the Maokou Formation, including paleokarst relief, sinkholes,caves, tower karst and its corresponding rocks (such as kaolinite, bauxite and ferruginous duricrust andcollapsed breccias, etc.). This paleokarst unconformity was in turn covered or infilled by the Emeishan basaltsand tuffs, suggesting that uplift and erosion occurred prior to the eruption of the ELIP. The extent of erosionof the Maokou Formation indicates the ELIP can be divided into three roughly concentric zones: the inner,intermediate, and outer zones. The paleokarst features on the top of Maokou Formation vary across the ELIP.In the inner zone, a likely sinkhole and an incision valley with 450 m relief in height are found. In theintermediate zone, various paleokarst landforms such as karst relief, sinkholes and tower karsts are welldeveloped. Some sinkholes that developed in the Qixia Formation below the Maokou Formation imply thatthe paleorelief is more than 350 m in height. In the outer zone, the paleokarstic surface is a paleo-weatheringlayer with minor karstification and development of caves at 10–50 m. This spatial variation of the paleokarstreflects variation of uplift height across the ELIP. The extent of minimal uplift is estimated to be at least450 m in the inner zone, 350 m in the intermediate zone, whereas uplift is minor (tens-50 m) in the outerzone. The magnitude and shape of the uplift is roughly consistent with that predicted by mantle plumemodels. The paleokarst was formed after the deposition of the Maokou Formation and the eruption of theEmeishan basalts at the end-Guadalupian and indicates a short duration of uplift. Thus this study lendsfurther support to domal uplift prior to the Emeishan flood volcanism, but also to the mantle plume initiationmodel for the generation of the ELIP.

: +86 020 85290130.

l rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Debate persists as to the existence of mantle plumes (Foulger andNatland, 2003; Depaolo and Manga, 2003). One of the importantsurface manifestations of a starting mantle plume is domal uplift ofthe Earth's surface (Rainbird and Ernst, 2001 and references therein;Campbell, 2007). Since plume-induced dynamic uplift should be pre-eruptive and basically transient, it is difficult to test for many largeigneous provinces (LIPs). The Emeishan large igneous province (ELIP)of southwest China is critical in this documentation, providing fieldevidence of pre-volcanic domal uplift in response to mantle plumeimpact on the lithosphere (He et al., 2003). However, Ukstins-Peateand Bryan (2008), on the basis of observations on hydromagmaticdeposits at Daqiao, cast doubt on an unconformity between the

Maokou limestone and the Emeishan basalts, and argued that theinitial eruption of the ELIP event was emplaced at or around sea leveland deny any pre-eruptive uplift. Since the Emeishan basalts generallyoverlay the Maokou Formation, a carbonate platform facies, thenature of the contact between both units is important to determine ifregional uplift occurred prior to eruption. For this purpose, a detailedcharacterization of the contact between the Emeishan basalts and theMaokou Formation has been carried out. Two positive aspects of suchan investigation are: (1) excellent exposures of Middle Permiancarbonate platform strata and the Emeishan basalts; and (2) largebodies of data on this contact are available in literature, due to theexploration for natural gas in the northeast of this LIP (e.g., Institute ofGeology, Chinese Academy of Sciences, 1979; Huang andWu, 1985; Liand Chen, 1993; Wang et al., 1994; Weng, 1996; Wang and Ji, 1997;Kang, 2000; Zhang and Liu, 2009) and ore deposits (e.g., Fan, 1993).Our study emphasizes the unconformity of the contact between theMaokou Formation and the Emeishan basalts, by providing newevidence of paleokarst on the top of the Maokou Formation.

2 B. He et al. / Lithos 119 (2010) 1–9

Furthermore, the spatial variation of the paleokarst is documentedand the cause of the different features across the ELIP is discussed inorder to determine if domal crustal uplift occurred prior to theEmeishan flood volcanism.

Paleokarst can be defined as ‘fossil’ karst, a remnant from aprevious period of karstification, and characterized by the presence ofancient, buried surfaces and deposits (Miller and Mitchell, 2003).Most documented paleokarsts are surficial, epikarstic phenomena andcavernous features (James and Choquette, 1988; Bosak et al., 1989;Wright, 1991). The proposedmethodology of Molina et al. (1999) andEsteban (1991) for palaeokarst analysis consists of the following: (a)the morphology of the surfaces, that is, karst landforms such assinkholes and caves; (b) the nature of the materials filling in thecavities or directly covering the irregular surfaces (such as kaolins,bauxites, calcretes and collapse breccias); (c) study and interpretationof exposure environments; and (d) analysis of underlying andoverlying rocks to recognize immediately pre- and post-unconformityevents. Among the above criteria, the karst landforms are the mostimportant features for the identification of paleokarst (James andChoquette, 1988; Wright, 1991; Molina et al., 1999). For this study weapply the methods described by Molina et al. (1999) and Esteban(1991) to examine the paleokarst features on top of the MaokouFormation.

2. Geological background and Permian stratigraphy

The ELIP in SW China, which consists of massive flood basalts andnumerous contemporaneous mafic and felsic intrusions, covers anarea of more than 2.5 105 km2 with a total thickness ranging from

Fig. 1. Schematic illustration of the geological features of the Emeishan large igneous provinseparate the inner, intermediate and outer zones, which are defined in terms of extent of e

several hundred meters to ~5 km (Fig. 1, Xu et al., 2001). TheEmeishan volcanic successions unconformably overlie the late MiddlePermian Maokou Formation (Fig. 2). The Maokou Formation has beenextensively studied in China due to its widespread distribution andextremely abundant fossils (Wang et al., 1994; Feng et al., 1997; ECS,2000). It mainly consists of medium bedded to massive biograinlimestones and biograin micritic limestones. Based on the fusulinidsassembly, the Maokou Formation was divided into three biostrati-graphic units (i.e., Neoschwagerina simplex, Neoschwagerina craticuli-fera-Afghanella schencki and Yabeina-Neomisellina zones) (ECS, 2000;He et al., 2003). The original thickness of the Maokou limestoneranges from 250 m to 500 m with an average of 350 m (Wang et al.,1994; Feng et al., 1997). The Permian strata in South China may bedivided, in ascending order, into Liangshan (Lower Permian), QixiaandMaokou (Middle Permian), andWujiaping and Changxing (UpperPermian) Formations (Fig. 2). In the view of rock types as well as fossilvariety of the Permian strata, its depositional environment wasinferred to be a typical carbonate platform (e.g., Wang et al., 1994;Feng et al., 1997; ECS, 2000). However, at the end of the MiddlePermian, the region experienced a rapid uplift, known as the Dongwumovement in Chinese geological community (Fig. 2) (ECS, 2000).

To characterize the crustal processes prior to the Emeishanvolcanism, He et al. (2003) correlated biostratigraphic units andthickness of the Maokou Formation. Results showed a systematicthinning of the strata beneath the Emeishan basalts. In terms of theextent of erosion of the Maokou Formation, the domal structureassociated with the Emeishan large igneous province can be dividedinto inner, intermediate, and outer zones (Fig. 1, He et al., 2003). Inthe following section, paleokarstic features on the top of the

ce and the locations of paleokarst surface on the top of Maokou formation. Dashed linesrosion of the Maokou Formation (He et al., 2003).

Fig. 2.Generalized Permian stratigraphy and division of fusulinid zones in SW China. Data compiled from ESC(2000); BGMRGP (1987), BGMRSP(1991). Age scheme is after Gradsteinet al (2004). R—regression; T—transgression.

3B. He et al. / Lithos 119 (2010) 1–9

Maokou Formation, with emphasis on its spatial variation across theELIP, are described respectively in the inner, intermediate and outerzones.

3. Paleokarst at the top of the Maokou Formation

3.1. Paleosinkhole (?) and incision valley (?) in the inner zone

Unlike the intermediate and outer zones where the MaokouFormation is covered by the Emeishan basalts, within the inner zonethe Emeishan basalts overlay Pre-Permian rocks in several localities(Fig. 3). One example is located at Hejia village, Panzhihua city,Sichuan province (Fig. 3A). At this locality, the Emeishan basaltsoverlay both the Maokou Formation and the Sinian dolomite. Sincethe Maokou Formation is interpreted to be a carbonate platform thatcovers the whole of SW China (e.g., Wang et al., 1994; Feng et al.,1997; ECS, 2000), this observation that basalts covered or infilled inSinian rocks in Fig. 3A may be interpreted to be a paleosinkhole.

Another example is located at Mishibozi village, Fuxin town,Mianning county, Sichuan province. The Emeishan basalts uncon-formably overlay Neoproterozic tonalite–trondhjemite–granodiorite(TTG) in the north of Mishibaozi village (Fig. 3B). TTG of this region ismetamorphosed into gneissic rocks, which are NNE distributed with asteep dip, whereas the Emeishan basalts gently cover the TTG series.Since the Permian, South China was covered by carbonate rocks

(Wang et al., 1994; Feng et al., 1997). The carbonate rocks covered anarea larger than the ELIP and suggest that the Permian sequenceswere totally eroded within the inner zone. The Mishibozi site may bean incision valley during the Maokou stage. This argument is alsosupported by canyon fillings, quartzose sandstones below theEmeishan basalts and above the Maokou formation in the west ofthis zone (He et al., 2006).

In the inner zone, theMaokou Formation is deeply eroded (He et al.,2003, 2006), with the thickness of the remnant Maokou limestonegenerally less than 50 m (compared with a thickness of 350 m for un-eroded standard stratigraphy). The weathering zone is also hard to findin the inner zone probably due to strong chemical dissolution anderosion in carbonate areas with its high elevation.

3.2. Paleokarst relief and landforms in the intermediate zone

3.2.1. Paelokarst relief in outcrops and bore-holesIn the intermediate zone, various paleokarst landforms and relief

were developed on the Maokou Formation (Fig. 1), as illustrated in anumber of outcrops and bore-holes (Fig. 4). The contact between theMaokou Formation and the Emeishan basalts is uneven or rugged. Therelief of the contact is usually several to tens of meters high. In someinstances, the relief of the contact can be as high as 30 m in outcropscale (Fig. 3B and G). An extreme example is found at Luoping county,Yunnan province, where the relief of the contact is up to 120 m high

Fig. 3. Schematic geological maps of Hejia and Mishibozi areas within the central ELIP. Locations are marked in Fig. 1. (A) The Emeishan basalts overlay both the Sinian dolomite andthe Permian limstone. (B) The Emeishan basalts overlay Neoproterzoic TTG. Pt represents Neoproterzoic TTG; Q represents Quaternary. Both figures show that Permian was totallyeroded in some localities within the central ELIP.

4 B. He et al. / Lithos 119 (2010) 1–9

and 5000 m in length (Fig. 3H). Based on detailed field work, it isdocumented that the relief of the contact varies from ~100 m, 20–50 m, several meters and to less than 1 m from the inner zonemarginsto outer zone margin across the intermediate zone (Institute ofGeology, Chinese Academy of Sciences, 1979). Since claystones andpaleosols such as kaolins, bauxites and ferruginous duricrust weredeveloped on the contact (Fig. 3E and F), which are interpreted asextreme weathering products of Maokou Formation, the relief of thecontact is not of primary sedimentary origin. In addition, beddings ofthe Emeishan basalts are truncated by the Maokou Formation in somelocalities (Fig. 3D and E), indicating likely topographic contrast. Itshould be emphasized that all the paleokarst relief is covered by theEmeishan basalts and tuff, suggesting that exposure and upliftoccurred before the eruption of the ELIP.

3.2.2. Sinkholes and tower karstMore than 60 sinkholes are found in Maokou and Qixia Formations

at Lunan county, 200 km east of Kunming City (Fig. 5A). Thesesinkholes, average ~1000 m2, are infilled with the Emeishan basalts,and were likely developed in a karst environment. Some collapseconglomerates are found at the base of those sinkholes (Institute ofGeology, Chinese Academy of Sciences, 1979). Magnetic explorationof these basalts indicated that the basalts extend deep into the caves(BGMRYP, 1990). Some sinkholes were developed on Qixia Formationin this region (Fig. 5A), implying that the saturated zone or the watertable is located below the layer, which is generally higher than sealevel. Given the average thickness of Maokou Formation of ~350 m,the extent of uplift in the intermediate zone is inferred to be greaterthan 350 m. Sinkholes on the top of Maokou Formation are also foundin the southern margin of Sichuan basin (Fig. 1; Institute of Geology,Chinese Academy of Sciences, 1979), where they are infilled bykaolins. These sinkholes are 1 to 5 m in diameter and 2 to 30 m indepth. Stone forest-style karst is observed at Dali, Yunnan province(Fig. 5B). They are 1000–3000 m in length, several hundred meters inwidth and up to 100 m in height.

3.2.3. The nature of the materials covering the paleokarstic surfacesClaystones, paleosols, clastic rocks and breccias etc. are developed

on some paleokarstic surface in the intermediate zone. Claystones andpaleosols such as kaolins, bauxites and ferruginous duricrust weredeveloped on rugged surfaces (Fig. 4C, E, and F), which are generallyinterpreted as extreme weathering products. Lignitic mudstones andsiltstones indicate terrestrial influx, distinct from platform carbonatefacies of underlying Maokou Formation. Rocks related to Maokoustage paleokarst have been analyzed in numerous sections duringstudy of the regional sequence stratigraphy or sedimentology of theSouthwest China (Xu et al., 1997). Bauxites, ferruginous mudstones,limestone breccias and coal seams with plant debris on the MaokouFormation imply a subaerial exposure. Collapse breccias and con-glomerates are also found on karstic surface of theMaokou Formation.One example is from Binchuan, where limestone breccias onpaleokarstic surface are cemented by basalts (Fig. 6).

3.3. Paleokarst of the Maokou Formation in the outer zone

3.3.1. Paleokarst features on the top of Maokou Formation in theouter zone

The contact between the Maokou Formation and the Emeishanbasalts in the outer zone consists mainly of paleoweathering crust andpaleosols withminor karstification. The paleoweathering zone containspyrite-bearing clay rocks, manganese and thin layers of coal andsiliceous rock. In some instances, kaolin and alumina ore deposits wereformed on the uneven erosional surface. Paleokarst relief and theerosion of the Maokou Formation are generally minor in this zone. TheEbian section reported by Lai et al (2008) is one example. The nature ofthe contact between theMaokou Formation and the Emeishan basalts iswell displayed along hillside outcrops (Section I in Fig. 1). The topsurface is undulatory, with a relief of 60 cm. The hollows of the karsticsurface are infilled by the paleosols. This is separated from the overlyingbasal basalt lavaflowby a thin succession of alternating coaly shales andcarbonaceous limestones.Xuet al. (1997)also reported several outcropsof the karstic contact between Maokou Formation and the Emeishan

Fig. 4. Paleokarst landforms and reliefs on the top of Maokou Formation. Locations are marked in Fig. 1. A—Daying Fuming county, Yunnan province; B—Lianshanpo, Panxian county,Guizhou province; C—Shangcang, Dali, Yunnan province; D—Tiechang, Qiubei county, Yannan province; E—Laochang, Huize county, Yunnan province; F—Liuhong, Meigu county,Sichuan province; G—Bainiuchang, Ningliang county, Yannan province; H—Tuoze, Luoping, Yunnan province.

5B. He et al. / Lithos 119 (2010) 1–9

basalts (Section J in Fig. 1) in which the top surface of the MaokouFormation has been eroded and marked by the paleokarst. In sometypical outcrops, there is a leached zone with minor karstification onwhich weathering crusts, paleosols and residual deposits There aremany erosional pits (several-ten centimeters in diameter) within theleached zone (Xu et al., 1997).

3.3.2. Cave system of the Maokou Formation from the exploration ofnature gas field

The Sichuan basin located in the northeast of the ELIP is aMesozoic to Cenozoic intracontinental basin (Fig. 1) rich in naturalgas reservoirs. In the south of this basin, Maokou stage paleokarst gas

reservoirs were discovered in the 1960s. Since then plenty of newinformation about this paleokarst was documented (e.g., Institute ofGeology, Chinese Academy of Sciences, 1979; Huang and Wu, 1985;Li and Chen, 1993; Weng, 1996; Kang, 2000). Based on bore-hole log,seismic response, rock core, bit cuttings, field investigation, twooccurrences of Maokou stage paleokarst have been revealed. Like thatdescribed for the intermediate zone, the most common case is thedevelopment of paleokarst on the top of the Maokou formation. Theother occurrence of paleokarst is a cave system located 10–50 mbelow the contact of Maokou formation and Emeishan basalts (e.g.,ISCAS, 1979; Huang and Wu, 1985; Li and Chen, 1993; Kang, 2000). Acave was found within the outer zone (Fig. 7). This cave is located

Fig. 5. (A) Schematic geological map of Lunan area showing paleosinkholes; (B) Schematic geological map of Xiangyun area showing tower karsts. Locations are marked in Fig. 1.

6 B. He et al. / Lithos 119 (2010) 1–9

14.5 m below top surface of the Maokou Formation, this cave isairfilled and enclosed by columnar calcite crystal (Fig. 7). Caveswithin outer zone are usually infilled by basalts, clays etc, andindicative of the paleokarst formation at Maokou stage. This suggeststhe minimal extent of uplift is about 50 m (the caves are generallyformed at close to the water table). The development of cavitiesrather than fractures is critical for gas reservoirs with high productionand big reserves.

4. Discussion

4.1. Nature of the contact between the Maokou Formation and theEmeishan basalts

Ukstins-Peate and Bryan (2008) pointed out two argumentsagainst paleokarst unconformity of Maokou Formation immediatelybeneath the Emeishan basalts. One is that the paleokarst may havemore recent origins; the other is that the uneven contact between theMaokou limestone and Emeishan flood basalts may reflect primaryrelief such as reefs. As we described earlier, the paleokarst on the topof Maokou Formation in SW China, including paleokarst reliefs,sinkholes, caves and its corresponding rocks are infilled and covered

Fig. 6. Schematic geological map of an air-filled cave enclosed by calcite crystals in outerzone of the ELIP. Location is marked in Fig. 1.

by the Emeishan basalts and/or tuff. There is no doubt that thepaleokarst was formed before eruption of the Emeishan basalts. As forthe second point, the depositional environment of the MaokouFormation is not a reef environment but on a shallow carbonateplatform(e.g., Wang et al., 1994; Feng et al., 1997; Xu et al., 1997).Although it is possible that primary relief of the contact arose fromphysiographic complexities in the carbonate platform, it is difficult toexplain large relief (120 m) within a short distance (5000 m) inFig. 4H. Especially, in the ELIP, some weathering products such askaolins, bauxites and terrestrial clastic rocks are found in numerousfield exposures; this indicates subaerial exposure environment of thissurface. In the identification of the paleokarst, Xu et al (1997) havecompared the rock fabrics to study the paleokarst within theirsequential and regional stratigraphic context because rock fabrics onthe Maokou Formation in this region are considered diagnostic of thedifferent subaerial exposure facies and environments.

Integration of karst landforms such as relief, sinkholes and caves,weathering products, collapse breccias and subaerial exposure faciesand environments, definitely indicates the presence of paleokarst onthe top of the Maokou Formation. All these, together with erosion ofbiostratigraphic strata (He et al., 2003) and catastrophic deposits (Heet al., 2006), confirm the unconformity nature for the contact betweenthe Emeishan basalts and the Maokou Formation. Actually, this karstunconformity is well documented in Chinese geological community(e.g., BGMRGP, 1987; BGMRSP, 1991; Wang et al., 1994; Xu et al.,1997; Feng et al., 1997; ECS, 2000).

4.2. Timing and duration of the paleokarst formation

Age of an unconformity surface associated with paleokarst canyield rough constraints on the formation timing of paleokarst (Bosaket al., 1989; Zhang and Liu, 2009). In other words, timing of paleokarstformation can be estimated by stratigraphic position. Generallyspeaking, the paleokarst was formed before the oldest stratigraphicsedimentation over the unconformity surface and after the latestcarbonate rock stratigraphic sedimentation under the unconformitysurface. In the Emeishan case, paleokarst relief, various paleokarstlandforms, caves and paleo-weathered crust and paleosols, wereburied and infilled by the Emeishan basalts and tuff. Thus there is nodoubt that the paleokarst was formed prior to the Emeishanvolcanism. The youngest carbonate rock under this paleokarsticsurface is the uppermost biostratigraphic unit (i.e., Yabeina-Neomi-sellina zones) of theMaokou Formation, suggesting that the formation

Fig. 7. Photo of conglomerates cemented by the Emeishan basalts on the paleokarstic surface. Location is marked in Fig. 1.

7B. He et al. / Lithos 119 (2010) 1–9

of paleokarst must have taken place at the end of the Maokou stage.Since the latest carbonate rock stratigraphic sedimentation under thispaleokarstic surface is the uppermost biostratigraphic unit (i.e.,Yabeina-Neomisellina zones) of the Maokou Formation, suggestingthat the formation of paleokarst must have taken place at the end ofMaokou stage. The end of Maokou stage can thus be taken as theupper limit of the karst formation. Recently, the age of the Emeishanbasalts is constrained at ~260 Ma (Zhou et al., 2002; He et al., 2007),coincident with the Middle-Late Permian boundary (Gradstein et al.,2004). This suggests that the plaeokarst terminated at Middle-LatePermian boundary. Thus the coverage of the paleokarst by theEmeishan basalts implies a very short duration of karst formation. Theassessment of the Middle Permian age of this karst formation in theELIP is in contrast with Ukstins-Peate and Bryan (2008) who arguedfor a more recent origin.

4.3. Uplift responsible for regional variation of the paleokarst

It is widely suggested that karst morphology is generallycontrolled by the water table (Florea et al., 2007). Sinkholes aregenerally formed in the vadose zone, where limestone dissolution isgreatest above the water table. Cave systems are formed in thephreatic zone close to the water table. Moreover, since the landscapedenudes to the lowest seasonal water table, topographic relief such asan incision valley are generally supposed to be a gauge to measureuplift height. Thus karstic morphologic features, especially cavesystems may constrain uplift height (Florea et al., 2007). Analyzingthe uplift height of the upper palaeokarst of the Maokou Formationacross the ELIP allows us to constrain the uplift variation.

In the inner zone, the paleokarst is mainly expressed as deeperosion of the Maokou Formation. In particular, the likely presence ofa sinkhole located at Hejia village and an incision valley (?) atMishibaozi implies ~450 m relief height (average thickness of Lowerand Middle Permian strata is ~450 m). Since sinkholes and incision

valleys are generally formed above the water table or sea level (Floreaet al., 2007), minimal uplift extent of ~450 m is constrained in theinner zone. At the same time, strong chemical dissolution in theuplifted area may be also unfavorable to develop weathering crust orpaleosols.

In the intermediate zone, various paleokarst landforms, relief andcaves are well developed on the Maokou Formation. Paleosols andclaystones deposited on some negative karstic surfaces such as sags.At Lunan area of the intermediate zone, the Emeishan basalts coveredand/or infilled on the Qixia Formation, indicating that minimal heightof the uplift may be more than 350 m (i.e., average thickness ofMaokou Formation in this region). Based on the paleokarst relief, it isdocumented that the relief of the contact varies from ~100 m, 50–20 m, several meters and to less than 1 m from the inner margin toouter margin across the intermediate zone (Institute of Geology,Chinese Academy of Sciences, 1979).

In contrast, the paeloweathering layers are well developed on thepaleokarstic surface in the outer zone. Minor paleokarst relief in theouter zonemay imply that the water systemwasmainly in the form ofsurface water. In other words, uplift in the outer zone is minor. Thecave system in the outer zone mainly occurs 10-50 m below theEmeishan basalts. This indicates that the magnitude of uplift in theouter zone is 10-50 m.

Fig. 8 shows a sketch of paleokarst on the top of Maokou Formationacross the ELIP. This spatial variation of the paleokarst reflects variationof magnitude of uplift across the ELIP. Extent of minimal uplift isestimated to be 450 m and 350 m within the inner and intermediatezones, whereas uplift is minor (tens–50 m) in the outer zone.

4.4. Implication for mantle plume in the involvement of the ELIP

The mantle plume model is accepted as a viable mechanism forthe generation of the ELIP. One of the important features of thismodel is domal uplift of the Earth's surface (Rainbird and Ernst,

Fig. 8. (A) Sketch showing paleokarst features on the top of Maokou formation across the ELIP. Note various paleokarst landforms; (B) Curve of uplift height across the ELIP. Upliftheights of different locations are estimated from text (Location 1 from sinkhole of Hejiao and incision valley of Meishibaozi; Location 2 from sinkholes of Lunan, Location 3 frompaleokarst relief of Luoping (Fig. 4H), Location 4 from cave system of southern Sichuan basin; Location 5 from Fig. 6).

8 B. He et al. / Lithos 119 (2010) 1–9

2001 and references therein; Campbell, 2007). Based on thecorrelation between the biostratigraphic units of the MaokouFormation and studies on catastrophic deposits, it is suggestedthat pre-volcanic domal uplift occurred prior to the flood volcanismin response to mantle plume impact on the lithosphere (He et al.,2003, 2006). However, Ukstins-Peate and Bryan (2008), on thebasis of observations on hydromagmatic deposits at Daqiao, raisedquestion as to the unconformity between the Maokou Formationand the Emeishan basalts, and argued that the initial eruption of theELIP event was emplaced at or around sea level. The data presentedin this paper provide further evidence in support of the unconfor-mity nature for the contact. This unconformity, also known asDongwu uplift or movement in the Chinese geosciences commu-nity, is characterized by paleokarst and weathering on the MaokouFormation. The fact that the fossil karst topography was infilled andburied by the Emeishan basalts indicates the formation of theunconformity immediately prior to the Emeishan volcanism (He etal., 2009). Unconformities require subaerial exposure which couldalso result from eustatic changes or crustal uplift. But spatialvariation of uplift in the ELIP indicates that this fossil karstunconformity was not caused by regression. As discussed above,the various extent of uplift across the ELIP can nicely account forthe spatial variation in the nature of the contact between theEmeishan basalts and the Maokou Formation. The magnitude andshape of the uplift are roughly consistent with that predicted bymantle plume models (Fig. 8B, Campbell and Griffiths, 1990;Griffiths and Campbell, 1991). More importantly, the nature ofthe unconformity varies systematically, a pattern can only beproduced by domal uplift (Griffiths and Campbell, 1991). On theother hand, the fact that the Emeishan basalts were emplaced atthe end-Maokou stage, is indicative of a very short duration of theuplift. In summary, the shape, magnitude and the brevity of theuplift agree well with the predictions of the plume hypothesis.

5. Conclusions

We provide evidence for paleokarst on top of the MaokouFormation. They include paleokarst relief, sinkholes, caves, towerkarst and its corresponding rocks (such as kaolinite, bauxite andferruginous duricrust and collapse breccias, etc). This paleokarsticsurface is covered or infilled by the Emeishan basalts and tuff,indicating that uplift and erosion occur before the eruption of the ELIPand the uplift duration is short. Spatial variation of the paleokarstacross the ELIP can be accounted for by domal uplift. The character-ization of the paleokarst on the top of Maokou Formation lendsfurther support to the mantle plume initiation model for thegeneration of the ELIP.

Acknowledgements

This research was supported by a grant from the National NaturalScience Foundation of China (40721063), National Basic ResearchProgram of China (973 Program) (2007CB411401) and the CAS/SAFEAInternational Partnership Program for Creative Research Teams(KZCX2-YW-T004). This is contribution No IS-XXX from GIGCAS.The author would like to thank the constructive reviews by LaiXulong, Greg Shellnutt and an anonymous referee.

References

Bosak, P., Ford, D.C., Glazek, J., Horacek, I., 1989. Paleokarst. A Systematic and RegionalReview. Elsevier, Amsterdam.

Bureau of geology and mineral resources of Guizhou province (BGMRGP), 1987.Regional Geology of Guizhou Province. Geological Publishing House, Beijing. (inChinese with English abstract).

Bureau of geology and mineral resources of Sichuan province (BGMRSP), 1991.Regional Geology of Sichuan Province. Geological Publishing House, Beijing. (inChinese with English abstract).

9B. He et al. / Lithos 119 (2010) 1–9

Bureau of geology and mineral resources of Yunnan province (BGMRYP), 1990.Regional Geology of Yunnan Province. Geological Publishing House, Beijing. (inChinese with English abstract).

Campbell, I.H., 2007. Testing the plume theory. Chemical Geology 241, 153–176.Campbell, I.H., Griffiths, R.W., 1990. Implications of mantle plume structure for the

evolution of flood basalts. Earth and Planetary Science Letters 99, 79–93.Depaolo, D., Manga, M., 2003. Deep origin of hotspots - is only seeing believing? Science

300, 920–921.Editing Committee of Stratigraphy (ECS), 2000. Permian in China. Geological Publishing

House, Beijing. (in Chinese with English abstract).Esteban, M., 1991. Paleokarst: practical applications. In: Wright, V.P., Esteban, M.,

Smart, P.L. (Eds.), Paleokarsts and Paleokarstic Reservoirs, 2. P.R.I.S. University ofReading. Occ. Publ. Series, pp. 89–119.

Fan, C.G., 1993. Karst mineral deposits in Yunnan province. Yunnan Geology 12,235–246 (in Chinese with English abstract).

Feng, Z.Z., Yang, Y.Q., Zin, Z.K., 1997. Lithofacies Paleogeography of Permian of SouthChina. Petroleum University Press, Beijing. (in Chinese with English abstract).

Florea, L.J., Vacher, H.L., Donahue, B.N.D., 2007. Quaternary cave levels in peninsularFlorida. Quaternary Science Reviews 26, 1344–1361.

Foulger, G.R., Natland, J.H., 2003. Is “hotspot” volcanism a consequence of platetectonics? Science 300, 921–922.

Gradstein, F.M., Ogg, J.G., Smith, A.G., Bleeker, W., Lourens, L.J., 2004. A new geologictime scale, with special reference to PreCambrian and Neogene. Episodes 22,83–99.

Griffiths, R.W., Campbell, I.H., 1991. Interaction of mantle plume heads with the Earth'ssurface and the onset of small-scale convection. Journal of Geophysical Research 96,18295–18310.

He, B., Xu, Y.G., Chung, S.L., Xiao, L., Wang, Y.M., 2003. Sedimentary evidence for a rapid,kilometer scale crustal doming prior to the eruption of the Emeishan flood basalts.Earth and Planetary Science Letters 213, 391–405.

He, B., Xu, Y.G., Wang, Y.M., Luo, Z.Y., 2006. Sedimentation and lithofaciespaleogeography in southwestern China before and after the Emeishan floodvolcanism: New insights into surface response to mantle plume activity. Journal ofGeology 114, 117–132.

He, B., Xu, Y.G., Huang, X.L., Luo, Z.Y., Shi, Y.R., Yang, Q.J., Yu, S.Y., 2007. Age and durationof the Emeishan flood volcanism, SW China: Geochemistry and SHRIMP zircon U-Pb dating of silicic ignimbrites, post-volcanic Xuanwei Formation and clay tuff atthe Chaotian section. Earth And Planetary Science Letters 306–323.

He, B., Xu, Y.G., Campbell, I., 2009. Pre-eruptive uplift in Emeishan? Nature Geoscience2, 531–532.

Huang, H.L., Wu, B.Q., 1985. Origin and distribution of gas reservoir formed by karst inthe south Sichuan. Karst of China 4, 308–313 (in Chinese with English abstract).

Institute of Geology, Chinese Academy of Sciences, 1979. Study of karst in China.Scientific Press, Beijing. in Chinese with English abstract.

James, N.P., Choquette, P.W., 1988. Paleokarst. Springer, New York.Kang, P.Q., 2000. Fossil Karst of Yangxing Series in Chishui Region. Guizhou Geology 17,

92–98 (in Chinese with English abstract).Lai, X.L., Wang, W., Wignall, P.B., Bond, D.P.G., Jiang, H.S., Ali, J.R., John, E.H., Sun, Y.D.,

2008. Palaeoenvironmental change during the end-Guadalupian (Permian) massextinction in Sichuan, China. Palaeogeography, Palaeoclimatology, Palaeoecology269 (1–2), 78–93.

Li, D.L., Chen, Z.Y., 1993. The Development of Paleo-karst in Yangxin Limestone in SouthSichuan. Journal of Huainan Minging Institute 13, 1–10 (in Chinese with Englishabstract).

Miller, D.J., Mitchell, S.F., 2003. Paleokarstic surfaces in the Upper Cretaceouslimestones of central Jamaica. Cretaceous Research 24, 119–128.

Molina, J.M., Ruiz-Ortiza, P.A., Vera, J.A., 1999. A review of polyphase karstification inextensional tectonic regimes: Jurassic and Cretaceous examples, Betic Cordillera,southern Spain. Sedimentary Geology 129, 71–84.

Rainbird, R.H., Ernst, R.E., 2001. The sedimentary record of mantle-plume uplift, in R.E.Ernst, and K.L. Buchan, eds., Mantle plumes: their identification through time,Boulder, Colorado. Geological Society of America Special paper 352, 227–245.

Ukstins-Peate, I., Bryan, S.E., 2008. Re-evaluating plume-induced uplift in the Emeishanlarge igneous province. Nature Geosciences 1, 625–629.

Wang, Y.S., Ji, Y.Z., 1997. The formation of dolomite and paleokarst of the LowerPermian Series in Sichuan basin. Journal of Chengdu University of Technology 24,8–16 in Chinese with English abstract.

Wang, L.T., Lu, Y.B., Zhao, S.J., Luo, J.H., 1994. Permian Lithofacies Paleogeography andMineralization in South China. Geological Publishing House, Beijing. (in Chinesewith English abstract).

Weng, J.T., 1996. Paleokarst development stage and resulting paleographic environ-ment in China. Carslogica Sinica 15, 74–81.

Wright, V.P., 1991. Paleokarst: types, recognition, controls and associations. In: Wright,V.P., Esteban, M., Smart, P.L. (Eds.), Paleokarsts and Paleokarstic Reservoirs, 2. P.R.I.S. Occ. Publ. Ser, pp. 89–119.

Xu, X.S., Liu, B.J., Zhao, Y.G., Lu, Y.F., 1997. Sequence stratigraphy and basin-mountaintransformation in the western margin of upper Yangtze landmass during thePermian to Triassic. Geological Publishing House, Beijing. (in Chinese with Englishabstract).

Xu, Y.G., Chung, S.L., Jahn, B.M., Wu, G.Y., 2001. Petrologic and geochemical constraintson the petrogenesis of Permian-Triassic Emeishan flood basalts in southwesternChina. Lithos 58, 145–168.

Zhang, B.M., Liu, J.J., 2009. Classification and characteristics of karst reservoirs in Chinaand related theories. Petroleum Exploration Development 36, 12–29.

Zhou, M.F., Malpas, J., Song, X.Y., Robinson, P.T., Sun, M., Kennedy, A.K., Lesher, C.M.,Keays, R.R., 2002. A temporal link between the Emeishan large igneous province(SW China) and the end-Guadalupianmass extinction. Earth And Planetary ScienceLetters 196, 113–122.