Late Eocene palaeogeography of the proto-Paratethys Sea in ...forth/publications/Bosboom_2015.pdf · Late Eocene palaeogeography of the proto-Paratethys Sea in Central Asia (NW China,

Late Eocene palaeogeography of the proto-Paratethys Sea in Central

Asia (NW China, southern Kyrgyzstan and SW Tajikistan)

RODERIC BOSBOOM1*, OLEG MANDIC2, GUILLAUME DUPONT-NIVET1,3,4,

JEAN-NOEL PROUST3, CHOLPONBEK ORMUKOV5 & JOVID AMINOV6

1Palaeomagnetic Laboratory Fort Hoofddijk, Faculty of Geosciences,

Utrecht University, Budapestlaan 17, 3584 CD Utrecht, The Netherlands2Geological–Palaeontological Department, Natural History Museum Vienna,

Burgring 7, A-1010 Wien, Austria3Geosciences Rennes, UMR 6118, Universite de Rennes 1, Campus de Beaulieu,

35042 Rennes Cedex, France4Key Laboratory of Orogenic Belts and Crustal Evolution, Ministry of Education

(Peking University), 100871 Beijing, China5Institute of Seismology: Kyrgyz Republic Bishkek, Asanbay 52/1,

720060, Bishkek, Kyrgyzstan6Institute of Geology, Earthquake Engineering and Seismology, 267 Ayni Street,

734053, Dushanbe, Tajikistan

*Corresponding author (e-mail: [email protected])

Abstract: The Cretaceous and Palaeogene sediments of the basins in Central Asia include theremnants of the easternmost extent of a vast shallow epicontinental sea, which extended acrossthe Eurasian continent before it retreated westwards and eventually isolated as the ParatethysSea. To improve understanding of its long-term palaeogeographical evolution, we complementthe well-constrained chronological framework of the Tarim Basin in China with stratigraphicrecords of the sea retreat from the Fergana Basin and the Alai Valley Basin in southern Kyrgyz-stan and the Afghan–Tajik Basin in SW Tajikistan. By lithostratigraphic analyses and identifi-cation of bivalve assemblages, this study establishes for the first time a clear and detailedregional correlation of Palaeogene marine strata across Central Asia, showing that the basinsshare a similar palaeogeographical evolution characterized by a long-term stepwise retreat punc-tuated by short-term shallow-marine incursions. Our correlation shows that the last two marineincursions recognized in the Tarim Basin can be traced westwards. The permanent disappear-ance of the sea from Central Asia probably occurred with limited diachroneity in the lateEocene, before the isolation of the Paratethys Sea, shifting the easternmost margin of the seahundreds of kilometres westwards and probably significantly reducing moisture supply to theAsian interior.

Central Asia has historically been known as animportant crossroad of cultures and the battle-ground of world powers, but this region is alsoof particular interest to geologists (Fig. 1). Thepalaeoenvironmental and palaeogeographical evol-ution of Central Asia since Eocene times resultsfrom the complex interplay between global climatechanges, regional tectonic uplift in response tothe collision of India and Arabia with Eurasia, andthe westward retreat of the shallow epicontinentalsea that extended across the Eurasian continent asfar east as the Tarim Basin in western China (e.g.Ramstein et al. 1997; An et al. 2001; Graham et al.

2005; Dupont-Nivet et al. 2007; Allen & Armstrong2008, 2012; Dupont-Nivet et al. 2008; Zhanget al. 2012). However, apart from the work of theSoviet field geologists (e.g. Vialov 1948; Mark-owsky 1959; Pomazkov 1972; Davidzon et al.1982; Dzhalilov et al. 1982), the sedimentary suc-cessions in its isolated basins that have recordedthe complicated geological history in Central Asiahave been relatively untouched. Here we focuson documenting the palaeogeographical extent andtiming of the sea retreat from Central Asia to bet-ter constrain its forcing mechanisms (i.e. eustacyv. tectonism) and palaeoenvironmental impacts, as

From: Brunet, M.-F., McCann, T. & Sobel, E. R. (eds) Geological Evolution of Central Asian Basins and the WesternTien Shan Range. Geological Society, London, Special Publications, 427, http://doi.org/10.1144/SP427.11# 2015 The Geological Society of London. For permissions: http://www.geolsoc.org.uk/permissions.Publishing disclaimer: www.geolsoc.org.uk/pub_ethics

at INIST-CNRS on September 25, 2015http://sp.lyellcollection.org/Downloaded from

mailto:[email protected]

http://sp.lyellcollection.org/

it may have served as an important moisture sourcefor the Asian interior (Ramstein et al. 1997; Zhanget al. 2007).

This sea was initially connected to the westernTethys before it retreated and, ultimately, separatedinto the Paratethys Sea (Baldi 1984; Rusu 1985;Dercourt et al. 1993; Robinson et al. 1996; Rogl1999; Popov et al. 2004; Schulz et al. 2005; Vincentet al. 2005; Allen & Armstrong 2008; Bosboomet al. 2011). We therefore refer to it as the proto-Paratethys Sea, although it is often named theTajik Sea, Turan Sea or Tarim Sea (e.g. Tanget al. 1992; Burtman & Molnar 1993). It supposedlyentered Central Asia from the west in the Early Cre-taceous and retreated westwards after five marineincursions (Tang et al. 1989; Burtman & Molnar1993; Lan & Wei 1995; Burtman 2000). Our pre-vious biomagnetostratigraphic study has accuratelydated the last two incursions into the Tarim Basin,showing that the retreat of the sea occurred step-wise and mainly in the late Eocene (Bosboomet al. 2011, 2014b, c). In order to evaluate thetiming and palaeogeography of the long-term step-wise sea retreat from Central Asia, this study aims

to correlate the poorly dated Palaeogene marinerecords further west in the Alai Valley Basin andthe Fergana Basin of southern Kyrgyzstan, and inthe Afghan–Tajik Basin of SW Tajikistan (Figs 1& 2) (e.g. Pomazkov 1972; Dzhalilov et al. 1982;Burtman 2000) to the well-dated chronostratigra-phic framework of the last two marine incursionsin the Tarim Basin. These results allow us to bet-ter constrain the westward stepwise sea retreatfrom Central Asia with respect to regional tecton-ism, Asian palaeoenvironmental change and globalsea-level variation.

Geological setting

Tectonism

The study area in southern Kyrgyzstan and SWTajikistan is bounded by the Pamir Mountainsin the south and the Tien Shan Mountains in thenorth. These thrust belts were activated in responseto the early Eocene Indo-Asia collision that oc-curred at around 50 Ma (e.g. Yin & Harrison2000; van Hinsbergen et al. 2012; Tripathy-Lang

Fig. 1. Satellite map of Central Asia showing the locations of the studied lithostratigraphic sections and the present-day extent and age of the marine sediments associated with the fourth and last fifth transgressions, based on data ofthis study and Bosboom et al. (2011, 2014b, c). The digital elevation data was downloaded from the online database ofthe CGIAR Consortium for Spatial Information (Jarvis et al. 2008). The locations of the Afghan–Tajik Basin, FerganaBasin, Alai Valley Basin, Tarim Basin and Xining Basin are shown on the inset (present-day coastal outlineobtained from GPlates 0.9.7.1).

R. BOSBOOM ET AL.



et al. 2013), resulting in regional tectonic upliftbetween the late Eocene and early Miocene (e.g.Thomas et al. 1994; Sobel & Dumitru 1997;Burtman 2000; Yang & Liu 2002; Yin et al. 2002;Sobel et al. 2006; Amidon & Hynek 2010; Ber-shaw et al. 2012; De Grave et al. 2012).

The Pamir forms the western extension of theTibetan Plateau. Here thrusting and exhumationoccurred mostly in the late Oligocene–early Mio-cene at approximately 25–18 Ma, based on sedi-mentological (Burtman 2000; Yin et al. 2002),stable isotope and provenance (Bershaw et al.

2012), thermochronological (Sobel & Dumitru1997; Jolivet et al. 2001; Amidon & Hynek 2010;Sobel et al. 2013), palaeomagnetic (Thomas et al.1994; Yin et al. 2002), and backstripping (Yang &Liu 2002) data. However, earlier uplift may haveinitiated some time in the middle Eocene but evi-dence is sparse and not well constrained (e.g.Jolivet et al. 2001; Yin et al. 2002; Amidon &Hynek 2010; Cowgill 2010).

The intracontinental Tien Shan is a Palaeozoicaccretionary orogen composed of multiple Pre-cambrian and Palaeozoic segments (e.g. De Grave

RishtanIsfara-Hanabad

Sumsar

TocharKushan

Hissarak-Shurysay

Sanglak

Shurysay

25

30

35

40

45

50

55

60

65

Olig

ocen

eP

aleo

cene

AG

E (M

a)

Ear

lyLa

teE

ocen

eE

arly

Mid

dle

Late

TARIM SEALEVEL & PALAEO-

GEOGRAPHYE

ML

5

4

3?

TARIM

EOT

GLOBALCLIMATE

ASIANARIDIFICATION

MECO

Last gypsum bedBiotic turnovers

Onset cyclicity

Step 1

Cooling event AGlacio-eustacy?

Cooling event CCAE-5

End lake systemEast Asian monsoon?

Upper Qimugen

Wulagen

Kalatar

Aertashi

Kezilouyi

Lower Qimugen

Bashibulake(5 members)

FERGANA& ALAI

AFGHAN-TAJIK

Bukhara

Turkestan

Alai

Akdzhar

Suzak

Tabakcha-Arukfan-Karatag

Beshkent

Jukar

Akdjar

Givar

MassagetKamoli

? ? Birth Paratethys?

Closure Turgai Strait

REGIONALTECTONICS

Pamir indentation?Tibetan uplift?

Indo-Asia collision~50 Ma

Uplift Pamir-Kunlun~25-18 Ma

Regressions Europe

Modern East Asian monsoon?

EECO

PETM

Fig. 2. Simplified lithobiostratigraphic correlation of the Fergana and Afghan–Tajik basins with the chronologicalframework of the marine incursions recognized in the SW Tarim Basin based on results of this study. The shaded areahighlights the formations corresponding to the fourth and fifth transgressions in the Tarim Basin, which have beenaccurately dated by integrated biomagnetostratigraphy (Bosboom et al. 2011, 2014b, c), allowing for temporalcomparison with the palaeogeographical evolution of the proto-Paratethys Sea (e.g. Baldi 1984; Rusu 1985; Dercourtet al. 1993; Proust & Hosu 1996; Robinson et al. 1996; Rogl 1999; Lopez-Blanco et al. 2000; Popov et al. 2004, 2008;Schulz et al. 2005; Vincent et al. 2005; Akhmetiev & Beniamovski 2006; Akhmetiev 2007; Lartaud 2007; Allen &Armstrong 2008; Johnson et al. 2009; Costa et al. 2010; Iakovleva & Heilmann-Clausen 2010; Dawber et al. 2011),the regional tectonic evolution of the Pamir, Kunlun Shan and Tien Shan (e.g. Hendrix et al. 1992; Yin & Harrison2000; Jolivet et al. 2001; Yang & Liu 2002; Yin et al. 2002; Robinson et al. 2003; Graham et al. 2005; Sobel et al.2006; Amidon & Hynek 2010; De Grave et al. 2012; van Hinsbergen et al. 2012; Yang et al. 2014), thepalaeoenvironmental changes recorded in the Asian interior (e.g. Meng et al. 1998; Sun & Wang 2005; Dupont-Nivetet al. 2007; Kraatz & Geisler 2010; Abels et al. 2011; Quan et al. 2011, 2012; Gomes-Rodrigues et al. 2012; Bosboomet al. 2014a) and global climate events (e.g. Pekar et al. 2002; Bohaty & Zachos 2003; Lyle et al. 2005; Milleret al. 2005; Tripati et al. 2005; Barker et al. 2007; Edgar et al. 2007; Katz et al. 2008; Kominz et al. 2008; Lear et al.2008; Tripati et al. 2008; Villa et al. 2008; Zachos et al. 2008; Bohaty et al. 2009; Edgar et al. 2010; Gasson et al. 2012),based on the geological timescale (Gradstein et al. 2012). Preliminary age estimates of the third transgression arebased on calcareous nannofossils (Zhong 1992), bivalves (Lan & Wei 1995), ostracods (Yang et al. 1995), dinoflagellatecysts (Mao & Norris 1988) and benthic foraminifera (Hao & Zeng 1984). The approximate relative changes in sea levelof each transgression in the Tarim Basin are shown by the thick-dotted line and are simply based on the reportedeastward extent of each incursion into the basin. The stratigraphic subdivision for southern Kyrgyzstan is based onPomazkov (1972) and for the Afghan–Tajik Basin on Dzhalilov et al. (1982). See Tables A1 & A2 for a comprehensiveoverview of the Palaeogene formations and corresponding lithostratigraphic descriptions, thicknesses and age estimatesfor southern Kyrgyzstan and the Afghan–Tajik Basin. EOT, Eocene–Oligocene transition; MECO, Middle EoceneClimatic Optimum; EECO, Early Eocene Climatic Optimum; PETM, Paleocene–Eocene Thermal Maximum.

PALAEOGEOGRAPHY OF THE PROTO-PARATETHYS SEA



et al. 2012). Thermochronological dating in theKashi Basin, the Tarim Basin and the Kyrgyz AlaiValley Basin indicates exhumation by reactiva-tion of the Late Palaeozoic thrust structures in theTien Shan that commenced near the Oligocene–Miocene boundary at approximately 24–22 Ma;southward propagation, however, did not reachthe Kashi Basin-bounding thrust until about 19 Ma(Sobel & Dumitru 1997; Sobel et al. 2006; DeGrave et al. 2012; Yang et al. 2014). Consequently,the molasse assemblages in Kyrgyzstan and Tajiki-stan eventually grade into coarse-grained sand-stones and conglomerates with mudstone interbedsof Miocene age (Markowsky 1959; Pomazkov1972; Coutand et al. 2002; Nikolaev 2002). TheseNeogene sediments have been weakly deformedby basinward thrusting and overloading of thePamir and Tien Shan, which is estimated to haveresulted in 300 km of total shortening (Burtman &Molnar 1993; Burtman 2000) and is ongoing upuntil today in response to the continuous northwardmovement of India into Eurasia.

This tectonic evolution had specific repercus-sions in the studied areas located in the Alai Val-ley Basin and Fergana Basin in Kyrgyzstan, andthe Afghan–Tajik Basin in Tajikistan (Fig. 1).

The east–west-trending Alai Valley is an intra-montane basin between the Trans-Alai and Alairanges that used to connect the present-day Afghan–Tajik and Tarim basins before the northwardindentation of the Pamir and the reactivation ofthe southern Tien Shan after the early EoceneIndo-Asia collision (Molnar & Tapponnier 1975;Hendrix et al. 1992; Burtman & Molnar 1993;Burtman 2000).

The Fergana Basin spreads mostly across easternUzbekistan, and has been marginally overthrustedby the Chatkal Range in the north (along theNorth Fergana Fault) and the Alai Range in thesouth (along the South Fergana Fault) in responseto the northward movement of India into Eurasia(Burtman & Molnar 1993; Burtman et al. 1996).Shortening in the Chatkal Range has been estimatedto be 60–100 km (Thomas et al. 1993; Zubovichet al. 2010). The NE margin of the basin isbounded by the major Talas-Fergana dextral strike-slip fault (Burtman et al. 1996). The basin has beenproposed to have formed in response to Permo-Triassic rifting of the Palaeozoic basement, and theoverlying sedimentary infill comprises Permian–Early Triassic sedimentary volcanic rocks, Jurassiccoals and Early Cretaceous continental deposits(Pomazkov 1972).

The Afghan–Tajik Basin spreads across easternUzbekistan, SW Tajikistan and NE Afghanistan,and has been marginally overthrusted by the TienShan in the north (Gissar Range) along the SouthGissar Fault, by the Pamir in the east (e.g. the

Darvaz, Trans-Alai and Peter the First ranges)along the major Darvaz Fault and by the HinduKush in the south along the Ishkashym Fault Zone(Brookfield & Hashmat 2001; Nikolaev 2002). Inthe west, the Afghan–Tajik Basin grades into theAmu-Darya Basin on the Turan Plate (Nikolaev2002). Basin development has been proposed tohave been initiated by Triassic rifting, and the sedi-mentary infill directly overlying the Palaeozoicbasement comprises Lower–Middle Jurassic coal-bearing deposits of both marine and continentalorigin, Upper Jurassic carbonate and halokineticsalt deposits of marine origin, and Early Cretaceouscontinental red beds (Markowsky 1959; Brookfield& Hashmat 2001; Nikolaev 2002).

Marine palaeogeography

Marine deposition in Central Asia initiated as theproto-Paratethys Sea invaded Central Asia throughthe Afghan–Tajik Basin in the Albian (Burtman &Molnar 1993; Burtman et al. 1996; Burtman 2000;Brookfield & Hashmat 2001; Nikolaev 2002),reaching the Tarim Basin in the Late Cretaceousthrough the Alai Valley Basin (Tang et al. 1992;Burtman & Molnar 1993; Burtman 2000). Herefive transgressions and regressions have been recog-nized, of which the third is considered the largest(Fig. 2) (Tang et al. 1989; Lan & Wei 1995; Burt-man et al. 1996; Burtman 2000). The subtidal faciesare characterized by shallow-marine carbonate-platform and tidal-flat deposits including light-coloured limestones and sandstones with typicalfossil assemblages, whereas the supratidal and inter-tidal facies comprise lagoonal and tidal-flat depositsincluding massive gypsum beds and reddish-browngypsiferous mudstones (Mao & Norris 1988; Tanget al. 1989, 1992). The last major retreat from thepalaeodepocentre in the SW Tarim Basin occurredat the Lutetian–Bartonian boundary at approxi-mately 41 Ma, according to integrated biomagne-tostratigraphic dating of marine deposits from thefourth transgression (base C18r: Bosboom et al.2011, 2014b). The fifth and last regression from thewesternmost margin of the Tarim Basin has beenassigned a latest Bartonian–early Priabonian ageby biostratigraphic constraints (near top C17n.2n–base C16n.2n: Bosboom et al. 2014c). These sea-retreat steps are concomitant with significantaridification in the Asian interior (Bosboom et al.2014a), East Asian monsoonal intensification(Quan et al. 2011), closure of the Turgai Strait inthe late Lutetian (Akhmetiev & Beniamovski2006; Akhmetiev 2007; Iakovleva & Heilmann-Clausen 2010) and various late Eocene regressionsreported from European basins (e.g. Proust &Hosu 1996; Lopez-Blanco et al. 2000; Lartaud2007; Costa et al. 2010; Dawber et al. 2011). A

R. BOSBOOM ET AL.



disconformity around the Eocene–Oligocene Tran-sition (EOT) near the palaeodepocentre shows thatthe Tarim Basin remained hydrologically part ofthe Tethyan realm up to at least the Oligocene,and had not yet been disconnected from theFergana and Afghan–Tajik basins to the west bythe tectonic closure of the Alai Valley Basin(Bosboom et al. 2014b).

The last regression in Central Asia is poorly con-strained to late Oligocene time (Markowsky 1959;Burtman et al. 1996; Burtman 2000). It is estimatedthat the complete marine sequence of Cretaceous–Palaeogene age in the Fergana Basin exceeds2000 m. In the northern Alai Valley Basin, this samesequence reaches a thickness in excess of 1000 m(Burtman & Molnar 1993; Burtman 2000). In theAfghan–Tajik Basin, the marine succession is divi-ded into two major marine sub-assemblages fromthe Albian to the Paleocene and from the Eoceneto the Oligocene, which together exceed 3000 mthickness in the centre of the basin (Burtman &Molnar 1993; Burtman 2000; Nikolaev 2002). Theoverlying molasse assemblages in both Kyrgyzstanand Tajikistan generally comprise late Oligocenefine-grained continental deposits with gypsum inter-calations (Markowsky 1959; Pomazkov 1972;Coutand et al. 2002; Nikolaev 2002).

Lithostratigraphy

Here we study the last two regional marine incur-sions recognized in several analysed sectionsthroughout the Alai Valley Basin, the FerganaBasin and the Afghan–Tajik Basin by describingand interpreting the general lithostratigraphy andlithofacies. This, and the review of previous lithos-tratigraphic descriptions and correlations (Pomaz-kov 1972; Dzhalilov et al. 1982; Burtman 2000),allow us to make a first-order lithostratigraphicregional correlation across these basins, which canultimately be linked to our previously establishedstratigraphic framework of the last two marineincursions in the Tarim Basin (Fig. 2) (Bosboomet al. 2011, 2014b, c). See Tables A1 & A2 in theAppendix for a comprehensive overview of theapplied stratigraphic nomenclature, and the cor-responding lithostratigraphic descriptions, thick-nesses and age estimates by Pomazkov (1972) andDzhalilov et al. (1982).

Sampled sections

In August 2011, five lithostratigraphic sectionswere studied in southern Kyrgyzstan to reconstructthe palaeogeographical development of the shal-low epicontinental sea covering Central Asia inthe Cretaceous and Palaeogene: the Northern Alai

Valley Composite section (39.68 N, 72.48 E) alongthe Alai Range (Tien Shan); the parallel sections ofDatka (39.98 N, 73.58 E) and Uch-Tobo (39.98 N,73.48 E) in the Alai Range (Tien Shan) betweenthe Alai Valley Basin and the Fergana Basin; andthe Tash-Kumyr (41.38 N, 72.28 E) and Ala-Buka(41.48 N, 71.48 E) sections in the Chatkal Range(Tien Shan) along the northern margin of theFergana Basin (Fig. 1). These sections are namedafter nearby villages or cities (except for the North-ern Alai Valley Composite section) and are exposedalong tributary streams of major rivers (except forthe upper part of the Tash-Kumyr section, whichis along the road from Tash-Kumyr to Jalal-Abad).These sections have been chosen for their subcon-tinuous exposure, lack of faulting and simple struc-tures, with a homoclinal dip ranging from 258 to658. The Northern Alai Valley has been studiedpreviously by Coutand et al. (2002) and Streckeret al. (2003), who focused, respectively, on lateCenozoic tectonic convergence and on active moun-tain front geomorphology. The Tash-Kumyr andAla-Buka sections were part of the palaeomagneticresearch by Thomas et al. (1993) on the generalCenozoic tectonic history of the Kyrgyz Tien Shan.

In the Afghan–Tajik Basin, three lithostrati-graphic sections covering Central Asia in the Cre-taceous and Palaeogene were studied in October2012: the Aksu section (38.118 N, 68.588 E) in theSW near the Uzbek border; the Childara–Shuldarasection (38.808 N, 70.358 E) on the western mar-gin of the Pamir Mountains; and the Kuhdara sec-tion (38.658 N, 68.888 E) near Dushanbe (Fig. 1).These sections are all named after nearby villagesor cities and are exposed along tributary streamsof major rivers. The strata are continuously exposedwith a homoclinal dip ranging from 358 to 808. TheAksu and Childara sections have been the focusof various palaeomagnetic studies (Bazhenov et al.1978; Klootwijk 1979; Bazhenov & Burtman1981; Thomas et al. 1994) directed at the generalCenozoic tectonic history of the Tien Shan andPamir.

The studied lithostratigraphic sections havebeen measured to decimetric precision (apart frompoorly exposed intervals) and are shown in Figures3 & 4. In general, the exposure is good, except infine-grained less-resistant intervals (particularly inthe Northern Alai Valley Composite section andthe sections in the Afghan–Tajik Basin). The suc-cessions encompass an alternation of regressiveand transgressive intervals (Fig. 5a). The regres-sive intervals are red-coloured oxidized successionscomprising playa evaporites, pedogenic calichebeds, floodplain mudstones, and fluvial channel silt-stones and sandstones with cross-bedding. Thetransgressive intervals are generally dominatedby shallow-marine, light-coloured (bioturbated)




Fig. 4. Regional lithobiostratigraphic correlation across the Afghan–Tajik Basin in SW Tajikistan. See Figure 3 for an explanation.

R.

BO

SB

OO

ME

TA

L.

at INIST

-CN

RS on Septem

ber 25, 2015http://sp.lyellcollection.org/

Dow

nloaded from


Fig. 5. Field photographs of sections, formations and sedimentological features from the Alai Valley Basin, FerganaBasin and Afghan–Tajik Basin. (a) Overview of the Datka section in the Alai Valley Basin, clearly showing thethird-order variations in sea level of the Palaeogene transgressions and regressions recognized in the Tarim Basin. At thetop is the continental Massaget Formation. The overall stratigraphic thickness is approximately 1500 m. (b) Oysterpackstone at the base of the Alai Formation in the Tash-Kumyr section in the Fergana Basin and (c) oyster samplescollected from that bed (hammer used for scale). (d) Overview of the last fifth marine incursion recognized at the top ofthe Shuldara section in the Afghan–Tajik Basin. Stratigraphic thickness of the greenish coloured marine beds isapproximately 200 m. (e) The gradual marine to continental transition of the last sea retreat as recorded in the Datkasection. Deeper marine green mudstones are overlain by shallow carbonates, which grade into red floodplain mudstoneswith playa evaporite interbeds at the base and fluvial sandstone interbeds at the top. At the top are the red alluvialconglomerates of the Massaget Formation. (f) Fluvial fine-grained sandstone channels interbedded in red floodplainmudstones at the base of the Shurysay Formation in the Aksu section in the Afghan–Tajik Basin (person used for scale).(g) Cross-bedded fluvial sandstones of the Shurysay Formation in the Datka section.




mudstones, marls, (oolithic) limestones and sand-stones rich in oysters and other bivalves.

The lower and penultimate incursion

The transgressive surface at the base of the lowermarine incursion is generally not well definedbecause of poor exposure but, in southern Kyrgyz-stan, it is marked by a rather abrupt shift from oxi-dized clastics to non-oxidized calcareous depositsrich in shallow-marine fauna. In both the Ferganaand Afghan–Tajik basins, the lithostratigraphy con-sists primarily of restricted shallow-marine depositscharacterized by light-coloured (bioturbated) mud-stones, marls and (oolithic) limestones and sand-stones rich in fish teeth, oysters and other bivalves(Fig. 5b–c), which is very similar to that of theKalatar and Wulagen formations of the fourthmarine incursion recognized in the Tarim Basin(Bosboom et al. 2011, 2014b). In the FerganaBasin and Alai Valley Basin, the end of this lowermarine incursion is characterized by the ratherabrupt transition into a relatively thin regressiveinterval of red-oxidized floodplain mudstones andfluvial fine-grained sandstone channels. The litho-logical characteristics of the deposits of this lowermarine incursion are typical of the Alai and Tur-kestan formations in southern Kyrgyzstan (Pomaz-kov 1972). In the Afghan–Tajik Basin, this lowermarine incursion consists of two transgression–regression sequences, which each range from open-marine subtidal to restricted lagoonal conditions andare separated by a clear unconformity (Fig. 4). Thiscontact is characterized by incised oxidized tidalchannels cutting into the restricted coastal marinesediments below at Aksu (at the 65 m level) andShuldara (at the 2 m level in the upper part), andby a sequence of fine-grained mica sandstonesdirectly overlying carbonate shoal oolithic grain-stones at Kuhdara (at the 65 m level). An intervalof red clays, siltstones and sandstones has beendescribed previously from the Jukar Formation,and the general lithofacies of the two sequencesare very typical of the Jukar, Beshkent and LowerTochar formations of the Afghan–Tajik Basin(Dzhalilov et al. 1982). The unconformity separ-ating the two sequences has not yet been identifiedat the corresponding stratigraphic level in the stud-ied sections in the SW Tarim Basin, which see-mingly shows a more simple third-order variationin sea level (Bosboom et al. 2011, 2014b). Thepossibility that the two stacked, marine sequencesin the Afghan–Tajik Basin correspond to the fourthand fifth marine incursions recognized in the SWTarim Basin appears unlikely. The stacked marinesequences, capped by the playa evaporite beds ofthe Lower Tochar Formation in the Afghan–TajikBasin, perfectly resemble the shallowing-upward

trend associated with the fourth marine incursionin the Bashibulake Mine section at the westernmostmargin of the Tarim Basin (Bosboom et al. 2014c).Hence, in the Alai Valley Basin, the Fergana Basinand the Afghan–Tajik Basin, the studied lowermarine incursion correlates to the fourth marineincursion of the Tarim Basin (Bosboom et al.2011, 2014b).

The upper and last incursion

The base of the upper marine incursion is onlyexposed in the Kyrgyz sections (the Fergana Basinand the Alai Valley Basin) and is, compared tothe previous incursion, marked by a more gradualchange from oxidized clastics to non-oxidizedcalcareous deposits rich in shallow-marine fauna.The marine sediments generally comprise red- andgreen-coloured (bioturbated) marls, and calcare-ous mudstones with occasional siltstone, sandstoneand limestone interbeds rich in shells (Fig. 5d),which is characteristic of a restricted subtidal–intertidal and low-energy depositional environment.In the Afghan–Tajik Basin, it is generally poorlyexposed and, at the Kuhdara section, it appears tobe completely absent. In the Alai Valley Basin andFergana Basin, the exposure is good but, owing tothe downcutting of the conglomeratic MioceneMassaget Formation, the ensuing regression onlyseems to have been fully preserved in the Datkaand Uch-Tobo sections. Here the marine sedimentsgrade into playa evaporite beds and red-colouredfluvial floodplain mudstones and sandstones, typi-cal of the Shurysay Formation (Fig. 5e–g). Theobserved green marly mudstone lithologies arecharacteristic of the marine Isfara, Hanabad andSumsar formations in southern Kyrgyzstan (Pomaz-kov 1972) and the Kushan and Sanglak forma-tions in the Afghan–Tajik Basin (Dzhalilov et al.1982), and correspond to the fine-grained marinemembers of the Bashibulake Formation of the fifthtransgression that are restricted to the westernmostmargin of the Tarim Basin (Bosboom et al. 2014c).

Synthesis

Our preliminary lithostratigraphic analyses indi-cate that the two last marine incursions can be cor-related across Central Asia. The transgressivesuccessions are thickest (reaching a total thicknessof over 600 m) and best recorded in the Datka andUch-Tobo sections in southern Kyrgyzstan (Fig.3). As in the Tarim Basin, the gradual character of thetwo studied regressions is typical of shallowing-upward cycles (or cyclothems) recorded in shallowepicontinental seas (Aigner et al. 1990; Reading2006). The same alternation of siliciclastic and car-bonate deposits is observed, although clastic input is

R. BOSBOOM ET AL.



generally higher than in the Tarim Basin. In south-ern Kyrgyzstan, medium- to coarse-grained sand-stone beds rich in clastic pebbles are predominantthroughout most sections, and the top of the lowerincursion into the Afghan–Tajik Basin is character-ized by frequent slumping, reworking, and mica-rich siltstones and fine sandstones (Figs 3 & 4).The mixed depositional environment suggests that,superimposed on the overall shallowing-upwardtrend, high-frequency fluctuations in relative sealevel, local climate and sediment supply result inlowstand/wet siliciclastic shedding and highstand/dry carbonate build-up (Budd & Harris 1990;Reading 2006).

Based on our findings, we correlated these lasttwo marine incursions in the Alai Valley Basin,Fergana Basin and Afghan–Tajik Basin to the lasttwo marine incursions recognized in the TarimBasin, which have been accurately dated by inte-grated biomagnetostratigraphy (Bosboom et al.2011, 2014b, c). This lithostratigraphic correlationwith the fourth and fifth transgressions of the SWTarim Basin is verified below by identification ofcollected bivalve specimens.

Bivalve analyses

The samples for bivalve analyses were collectedfrom representative marine beds throughout thestudied sections (Figs 3, 4, 6 & 7; Table 1) in orderto verify our lithostratigraphic correlations, and toconstrain the palaeogeography and palaeoenviron-ment of the stepwise sea retreat from Central Asia.All identified species are documented in Figure 6.Their distribution in the studied samples is indi-cated in Table 1. The correlations for the Alai Val-ley Basin and the Fergana Basin are partly basedon studies from the Afghan–Tajik Basin, of whichthe formation names have been transferred to thestratigraphic nomenclature applied in southern Kyr-gyzstan according to the stratigraphic scheme ofBurtman (2000).

Bivalves from the Alai Valley Basin and the

Fergana Basin (southern Kyrgyzstan)

The Alai section. Sample AL11-S03 yielded well-preserved specimens of Sokolowia buhsii (Gre-wingk 1853) (Fig. 6a: 6–8, 10), confirming thelithostratigraphic correlation with the TurkestanFormation (Vialov 1948). Sokolowia buhsii (Gre-wingk 1853) is typical of the Wulagen Formation,which records the fourth marine incursion into theSW Tarim Basin (Lan 1997) and has been accu-rately dated by integrated biomagnetostratigra-phy as Lutetian (Bosboom et al. 2014b). It pointsto normal saline, warm and turbulent subtidal

shallow-marine water conditions and shows extra-ordinary wide dispersal from the Tarim to the Tran-sylvanian basins (Grewingk 1853; Vialov 1937;Berizzi Quarto di Palo 1970; Rusu et al. 2004).For a detailed review of the palaeoecological, strati-graphic and palaeogeographical significance ofSokolowia buhsii (Grewingk 1853) see Bosboomet al. (2011). From sample AL11-S02, only indeter-minate gastropods with no biostratigraphic valuewere retrieved. Sample AL11-S01, taken 350 mbelow, revealed a bivalve assemblage of Upper Cen-omanian age (Fig. 6a: 6–8).

The Datka and Uch-Tobo sections. Based on ourlithostratigraphic correlation, the two samplesUT11-S01 and UT11-S02 (Fig. 6a: 9; Fig. 6b: 9,17, 18) would originate from the Alai–Turkestanformations, which is perfectly in line with the identi-fication of Sokolowia buhsii (Grewingk 1853). Thespecimens from sample UT11-S02 somewhat remi-niscent of Ostrea (Turkostrea) cizancourti (Cox1938), which is typical of the Alai Formation(Berizzi Quarto di Palo 1970). However, carefulcomparison of the 10 available specimens shows itis, in fact, a more slender and small-sized morpho-type of Sokolowia buhsii (Grewingk 1853) reportedfrom the Turkestan Formation of northern Afgha-nistan by Berizzi Quarto di Palo (1970). See theprevious subsection on ‘The Alai section’ for itspalaeoenvironmental significance and palaeogeo-graphical distribution.

The specimens of sample UT11-B02 were col-lected from various shell beds at the base of theupper transgression, which has been lithostrati-graphically correlated to the top of the RishtanFormation and the base of the Isfara Formation. Pla-tygena asiatica (Romanovskiy 1879) and Ferganeaferganensis (Romanovskiy 1879), found therein,represent index fossils of two different stratigraphiclevels (e.g. Berizzi Quarto di Palo 1970). Platygenaasiatica (Romanovskiy 1879) confirms correla-tion with the Rishtan Formation of the FerganaBasin (Vialov 1937; Berizzi Quarto di Palo 1970).Further east, Platygena is the index fossil of theSecond and lowermost Third Member of theBashibulake Formation (Lan 1997). This formationrecords the fifth marine incursion in the TarimBasin, and has been biostratigraphically constrainedin age between the latest Bartonian and early Priabo-nian (Bosboom et al. 2014c). Platygena asiatica(Romanovskiy 1879) appears to have a wide dis-tribution based on reports from Eocene strata ofLibya in North Africa (Cox 1962). This species pro-duces remarkable thick shells with sizes of up to15 cm and was probably a sediment recliner onsandy bottoms in fully marine, shallow subtidal–lower intertidal environments (Lan 1997). How-ever, in the Fergana Basin, Ferganea ferganensis




Fig. 6. (a) Plates showing the most significant macrofossil species identified (see also Table 1). 1. Flemingostrea ?hemiglobosa (Romanovskiy 1884) – DS12-S01. 2. Ostrea (Turkostrea) strictiplicata (Raulin & Delbos 1855) –AS12-S01. 3 & 4. Ostrea (Turkostrea) strictiplicata (Raulin & Delbos 1855) – DS12-S02. 5. Ostrea (Turkostrea)afganica (Vialov 1938) – AS12-S01. 6. Plicatula sp. – AL11-S01. 7. Costagyra olisiponensis (Sharpe 1850) –AL11-S01. 8. Curvostrea rouvillei (Conquand 1862) – AL11-S01. 9. Sokolowia buhsii (Grewingk 1853) – UT11-S01.10. Sokolowia buhsii (Grewingk 1853) – AL11-S03. 11. Platygena asiatica (Romanovskiy 1879) – AB11-S01.(b) Continued from (a). 1 & 2. Cubitostrea plicata (Solander 1766) – AB11-S02. 3–8. Ferganea ferganensis(Romanovskiy 1879) – AB11-S01. 9. Platygena asiatica (Romanovskiy 1879) – UT11-B02. 10. Flemingostrea ?hemiglobosa (Romanovskiy 1884) – TK11-S04. 11. Sokolowia buhsii (Grewingk 1853) – TK11-S02. 12. Crasssatella(Landinia) ustjurtensis (Ilyina 1955) – TK11-S03. 13 & 14. Glossus (Aralocardia) eichwaldiana (Romanovskiy1890) – TK11-S03. 15–16. Sokolowia buhsii (Grewingk 1853) – UT11-S02. 17 & 18. Ferganea ferganensis(Romanovskiy 1879) – UT11-B02.

R. BOSBOOM ET AL.



(Romanovskiy 1879) has previously only beenreported from the Sumsar Formation (Salibaev1972). In the SW Tarim Basin, it also shows a dis-tinctly different stratigraphic distribution from Pla-tygena asiatica (Romanovskiy 1879), marking theuppermost Third and Fourth Members of the Bashi-bulake Formation (Lan 1997). Ferganea ferganen-sis (Romanovskiy 1879) was endemic to Central

Asia and, in all probability, was an euryhalinespecies well adapted to environmental fluctuationscommon in palaeogeographically restricted basins(Lan 1997).

The macrofossil analyses are in perfect agree-ment with our previous lithostratigraphic corre-lation with the last and fifth transgression in theSW Tarim Basin. Beyond that, the co-occurrence

Fig. 6. Continued.




Table 1. Macrofossil content and corresponding stratigraphic interpretation of examined samples

Molluscs Kyrgyzstan–Tajikistan Taxa

Fourthtransgression

Fifth transgression Non-agediagnostic

Basin Section Code SampleStratigraphic

level (m) Formation

Age (afterBurtman

2000)

Fle

min

gost

rea

?hem

iglo

bosa

(Rom

anovsk

iy1884)

Ost

rea

(Turk

ost

rea)

afg

anic

a(V

ialo

v1938)

Ost

rea

(Turk

ost

rea)

stri

ctip

lica

ta(R

auli

n&

Del

bos

1855)

Soko

low

iabuhsi

i(G

rew

ingk,

1853)

Cra

sssa

tell

a(L

andin

ia)

ust

jurt

ensi

s(I

lyin

a1955)

Cubit

ost

rea

pli

cata

(Sola

nder

1766)

Fer

ganea

ferg

anen

sis

(Rom

anovsk

iy1879)

Glo

ssu

s(A

ralo

card

ia)

mic

a(O

vec

hkin

1954)

Pla

tyg

ena

asi

ati

ca(R

om

anovsk

iy1879)

Cord

iopsi

ssp

.

Gas

tropoda

indet

.

Gly

cym

eris

sp.

Alai Valley Alai Valley AL11 S02 +5 – – xS03 +170 Turkestan Bartonian x

Datka UT11 S01 +205 Turkestan Bartonian xUch-Tobo S02 +225 Turkestan Bartonian x

B02 + 430 Rishtan-Sumsar Priabonian x xFergana Tash-Kumyr TK11 S01 670.0 – – x

S02 605.5 Turkestan Bartonian xS03 624.0 Isfara-Hanabad Priabonian x xS04 500.5 Suzak Ypresian xS05 490.5 Suzak Ypresian x

Ala-Buka AB11 S01 +140 Rishtan-Sumsar Priabonian x xS02 +165 Sumsar Priabonian x

Afghan–Tajik

Kuhdara DS12 S01 0.6 Givar Ypresian x

S02 31.0 Jukar Lutetian xS03 25.6 – – x x

Aksu AS12 S01 20.8 Jukar Lutetian x x

R.

BO

SB

OO

ME

TA

L.

at INIST

-CN

RS on Septem

ber 25, 2015http://sp.lyellcollection.org/

Dow

nloaded from


of Platygena and Ferganea in the stratigraphicinterval at the base of the upper transgressionsuggests that the Ferganea species have a somewhatbroader stratigraphic occurrence in the study regionthan previously reported, extending downwards tothe base of the Isfara Formation.

The Tash-Kumyr section. Samples TK11-S05 andTK11-S04 comprise shells of Flemingostrea ?hemiglobosa (Romanovskiy 1884) (Fig. 6b: 10–16) confirming correlation with the Suzak For-mation (Vialov 1948), which has been assigned anYpresian age (Burtman 2000). This species hasalso been reported from the middle part of theQimugen Formation in the Tarim Basin, whichwas deposited during the regression following thethird marine incursion (Lan 1997), poorly con-strained in age between the late Paleocene andearly Eocene (Hao & Zeng 1984; Mao & Norris1988; Tang et al. 1989; Zhong 1992; Lan & Wei1995; Yang et al. 1995; Burtman 2000). Accordingto Lan (1997), this extraordinary thick-walled sedi-ment incliner represents the name-bearing speciesof the Flemingostrea–Panopea Assemblage Zone,marking a fully marine, subtidal setting with rela-tively warm seawater.

The presence of Sokolowia buhsii (Grewingk1853) in sample TK11-S02 confirms the correla-tion with the Turkestan Formation (Vialov 1948).See the earlier subsection on ‘The Alai section’for its palaeoenvironmental and palaeogeographicaldistribution.

Glossus (Aralocardia) mica (Ovechkin 1954)and Crasssatella (Landinia) ustjurtensis (Ilyina1955) in sample TK11-S03 correlate to the upperThird Member of the latest Bartonian–early Priabo-nian Bashibulake Formation in the SW TarimBasin (Lan 1997), and thus correlate with theFerganea-bearing horizon of the previous section.Specimens from the Tarim Basin identified byLan & Wei (1995) as Glossus (Aralocardia) eich-waldiana (Romanovskiy 1890) are identical toGlossus (Aralocardia) mica (Ovechkin 1954), andrepresent an erroneous identification based on itssmaller size and distally increasing prominence ofcommarginal folds. Glossus (Aralocardia) mica(Ovechkin 1954), originally described from theupper part of the Priabonian Chegan Formation inwestern Kazakhstan (Ovechkin 1954; Krasheninni-kov & Akhmetiev 1998), has been previously con-sidered as restricted to the Hanabad Formation inthe Afghan–Tajik Basin (Salibaev 1972). Thisstudy documents their presence near the base ofthe Isfara Formation. Crasssatella (Landinia) ust-jurtensis (Ilyina 1955) was originally recordedfrom the Priabonian of the Ustyurt Plateau inCentral Asia (Lan & Wei 1995). This distributionof the infaunal glossids and crassatelids points to

well-established palaeogeographical connectionsto the west, as well as to the east, during deposi-tion of the Isfara Formation. These species arecharacteristic of low-energy, subtidal depositionalsettings. From TK11-S01, only two Cordiopsis sp.steinkerns are available, which have no biostrati-graphic value.

The Ala-Buka section. Based on our lithostrati-graphic correlation, the collected samples originatefrom the Rishtan, Isfara and Hanabad formations(Fig. 6a: 11; Fig. 6b: 1–8). Sample AB11-S01 wascollected from shell beds at the same stratigraphicinterval at the base of the upper transgression as sam-ple UT-B02 of the Datka and Uch-Tobo sections(see the subsection on ‘The Datka and Uch-Tobosections’) and has the same remarkable co-occur-rence of Platygena asiatica (Romanovskiy 1879)and Ferganea ferganensis (Romanovskiy 1879).Platygena is an index species of the Rishtan For-mation and Ferganea of the Isfara, Hanabad andSumsar formations (Salibaev 1972), and both corre-spond to the marine members of the BashibulakeFormation representing the fifth marine incursionin the Tarim Basin (Lan 1997). As the transgressivesequence in the Ala-Buka section is condensedand constitutes only few tens of metres, their co-occurrence in the sampled stratigraphic intervalfits with their previously established stratigraphicoccurrence.

Cubitostrea plicata (Solander 1766) in AB11-S02 confirms the correlation with the Sumsar For-mation (Salibaev 1972). This species occurs onlyin the Fourth Member of the Bashibulake Formationin the Tarim Basin (Lan & Wei 1995), and is asmall-sized, shallow-water dweller, well adaptedto turbulent water and environmental perturbation(Lan 1997). This species has a wide distributionand is known, for example, from Priabonian depos-its of Bulgaria (Karagiuleva 1964).

Bivalves from the Afghan–Tajik Basin

(SW Tajikistan)

The Kuhdara section. Three well-preserved valvesof Flemingostrea ? hemiglobosa (Romanovskiy1884) (Fig. 6b: 1, 3, 4) are available from sampleDS12-S01, which supports our correlation withthe Givar Formation of Ypresian age (Dzhalilovet al. 1982; Burtman 2000) and corresponds to thelate Paleocene–early Eocene Qimugen Formation,which represents the regression following the thirdmarine incursion recognized in the SW depressionof the Tarim Basin (Lan 1997). Classification ofthe three Flemingostrea specimens as very largeand flat Rupelian Platygena asiatica (Romanovskiy1879) can be excluded based on the highly globose




left-valve and the absence of a narrow dorsallyelongated interior shell surface or glossy lamellaewith Camptonectes-type macrosculpture on the shellexterior. See the earlier subsection on ‘the Tash-Kumyr section’ for the palaeoenvironmental andpalaeogeographical distribution of Flemingostrea ?hemiglobosa (Romanovskiy 1884). Sample DS11-S03 merely contains steinkerns of articulated infau-nal bivalves Cordiopsis sp. and Glycymeris sp.,which have limited biostratigraphic significanceand point to a subtidal marine depositional setting(Baldi 1973). The occurrence of Ostrea (Turkos-trea) strictiplicata (Raulin & Delbos 1855) insample DS12-S02 confirms the lithostratigraphicposition within the Jukar Formation (Dzhalilovet al. 1982). The specimens are identical by right-valve and by left-valve to specimens illustrated inStenzel (1971). Their possible classification withthe Priabonian Cubitostrea plicata (Solander 1766)is excluded, as this species is larger in size and has adistinctly larger number of ribs on the left-valve.The Last Occurrence Datum (LOD) of Turkostreais in the Alai Formation (Stenzel 1971), and it isvery abundant in the lower and middle parts of theLutetian Kalatar Formation, recording the trans-gression of the fourth incursion in the Tarim Basin(Lan 1997; Bosboom et al. 2014b). The ostreidsare common in number and build bioherms (orreefs) in a lower subtidal–shallow subtidal sett-ing (Lan 1997). Ostrea (Turkostrea) strictiplicata(¼ Ostrea turkestanensis Romanovskiy 1880) hasalso been reported from the Persian Gulf, beingcommon in the middle Eocene of the BahrainIsland (Cox 1936).

The Aksu section. Correlation of the base of thesection to the Jukar Formation is well supportedby the oyster species from sample AS12-S01, ident-ified as Ostrea (Turkostrea) afganica (Vialov 1938)and as Ostrea (Turkostrea) strictiplicata (Raulin &Delbos 1855) (Fig. 6a: 2, 5) based on data fromthe Afghan–Tajik Basin by Vialov (1948). Thesespecies occur in the Alai Formation in northernAfghanistan (Cox 1938; Berizzi Quarto di Palo1970). In the Tarim Basin, the species are com-mon from the latest Paleocene to the middleEocene in the Ostrea (Turkostrea) afganica–Soko-lowia orientalis Assemblage Zone of the UpperQimugen Formation and in the Ostrea (Turko-strea) strictiplicata–Ostrea (Turkostrea) cizan-courti Assemblage Zone of the Kalatar Formation,recording the regression following the third incur-sion and the transgression of the fourth incursion(Lan 1997). These species formed bioherms inlower intertidal and shallow subtidal, fully marinewaters (Lan 1997). Whereas Ostrea (Turkostrea)afganica (Vialov 1938) is restricted to CentralAsia (Berizzi Quarto di Palo 1970), Ostrea

(Turkostrea) strictiplicata (Raulin & Delbos 1855)has a much wider distribution, from the TarimBasin up to the Bahrain Island in the Persian Gulf(Cox 1936).

Synthesis

The identified bivalve specimens largely confirmour lithostratigraphic correlation, showing that theSW Tarim Basin, Fergana Basin, Alai ValleyBasin and Afghan–Tajik Basin share a similar geo-logical history of shallow-marine third-order trans-gressions and regressions. The regional bivalveanalyses show that the two studied incursions areeach characterized by their own assemblage anddepositional environment, and support correlationof the two studied incursion to the recognized andwell-dated fourth and fifth incursions in the TarimBasin. In addition, the bivalve species collected atthe base of the Tash-Kumyr section in the FerganaBasin and at the base of the Kuhdara section in theAfghan–Tajik Basin correlate to the regressiveinterval preceding the fourth marine incursion.

Discussion

Based on our new litho- and biostratigraphic con-straints, this study establishes for the first time aclear and detailed regional stratigraphic correlationbetween the basins across Central Asia from SWTajikistan to NW China (Figs 2 & 7). The fourthand fifth marine incursions are each characterizedby distinct lithofacies and assemblages of bivalves,which are very similar to the lithofacies and ident-ified assemblages of the corresponding marineincursions recorded in the Eocene sedimentary suc-cessions of the Tarim Basin (Fig. 7). In the Afghan–Tajik Basin, no bivalve taxa representing the fifthmarine incursion have been recovered: however,the corresponding strata correlate lithostratigraphi-cally to bivalve-bearing deposits of the fifth incur-sion in the Alai Valley Basin, the Fergana Basinand the Tarim Basin. Hence, our correlation con-firms that, during both the fourth and fifth marineincursions, the studied basins became unified asthe shallow-marine epicontinental sea transgressedacross Central Asia. The sedimentary successionsof SW Tajikistan and southern Kyrgyzstan cannow thus be convincingly linked to the previouslyestablished chronological framework of the lasttwo marine incursions in the SW Tarim Basin,which have previously been dated as Lutetian andlatest Bartonian–early Priabonian, respectively(Bosboom et al. 2014b, c). This allows us to eva-luate the palaeoenvironment, palaeogeography,controlling mechanisms and palaeoenvironmentalimpacts of the long-term retreat of the proto-Paratethys Sea from Central Asia (Figs 1 & 8).

R. BOSBOOM ET AL.



Palaeoenvironment and palaeogeography

As in the Tarim Basin, the marine depositionalenvironment is generally characterized by fullymarine, shallow-water, near-shore conditions. Warm-water carbonate platforms with shoals and biohermsof reef-building oysters are typical of the fourthtransgression. Our preliminary lithostratigraphiccorrelation of this transgression shows that this

incursion is more complex in the Afghan–TajikBasin, where it comprises two stacked transgres-sion–regression sequences separated by a pro-minent unconformity. The deposits of the fifthand last transgression primarily include subtidal–intertidal low-energy mudstones. In general, thebivalve species recorded in both transgressionshave a wide palaeogeographical distribution as faras the Atlantic and the Indian oceans. However, the

Fig. 8. Preliminary palaeogeographical maps displaying the stepwise retreat of the proto-Paratethys Sea from CentralAsia in the middle–late Eocene (a) and the isolation of the Paratethys Sea in the early Oligocene (b). Note that thecoastlines are very approximate and based on Dercourt et al. (1993), Popov et al. (2004, 2010), Bosboom et al. (2011,2014b, c) and the findings of this study. The suggested changes in palaeogeography shown in the Turgai Strait andSiberian Sea are based on Akhmetiev & Beniamovski (2006), Akhmetiev (2007) and Iakovleva & Heilmann-Clausen(2010), and from an extrapolation of the results of this study. Plate boundaries were obtained from GPlates 1.2.0 for 38and 32 Ma. The stepwise retreat is time equivalent with significant aridification steps reported from the Xining Basin(Abels et al. 2011; Bosboom et al. 2014a).




present-day extent of the marine sediments of thefifth transgression is more limited, as shown bytheir absence at the palaeodepocentre of the SWTarim Basin and at the Kuhdara section in theAfghan–Tajik Basin. At the top of this final mar-ine succession, the bivalve species Ferganea,which is an endemic genus of the easternmost basinsof the proto-Paratethys originating within the fifthtransgression, reflects the last marine pulse in thebasins in Central Asia before their complete isolation(Figs 1 & 7).

As the sea entered and withdrew relativelyrapidly across the SW Tarim Basin during thefourth marine incursion (Bosboom et al. 2014c)and as water depths are relatively shallow in theseepicontinental basins (Bosboom et al. 2011,2014c; this study), we concluded that the diachrone-ity associated with transgressions and regressionsacross Central Asia was limited. Accordingly, weassume that the last marine deposits in CentralAsia, which have been correlated to the latest Barto-nian–early Priabonian fifth transgression of theTarim Basin, are probably of late Eocene age.This correlation with the chronostratigraphic frame-work of the Tarim Basin implies that the base ofthe regressive Hissarak (in the Afghan–Tajik Basin)and Shurysay formations (in the Alai Valley andFergana basins) overlying the last fully marinedeposits are probably of late Eocene age, whichis much older than the Oligocene age previouslyassigned (Fig. 2) (Markowsky 1959; Pomazkov1972; Burtman 2000; Nikolaev 2002). Our correla-tion indicates that this late Eocene regression marksthe end of the last major regional marine incursionand does not support a subsequent early Oligocenetransgression, as suggested by Burtman (2000).

This new regional stratigraphic correlationallows us to update the palaeogeographical mapsof the proto-Paratethys Sea (e.g. Dercourt et al.1993; Rogl 1999; Popov et al. 2004, 2010) withthe previously relatively unknown Central Asianpalaeogeography of the two late Eocene marineincursions, as shown in Figure 8. Unless youngermarine deposits have been eroded, our correlationof the last marine sediments suggests that the perma-nent disappearance of the sea from the studiedbasins would have occurred in the late Eocene.This final retreat of the proto-Paratethys Sea fromCentral Asia during late Eocene times fits with thepalaeogeographical maps of Popov et al. (2004,2010), which show that the Turan Sea largelycovered Central Asia in the late Eocene, while, inthe early Oligocene, the easternmost marginextended no further east than the area between thepresent-day East Aral and Syr-Darya basins. Thisis in agreement with the absence of organic-richand mud-prone deposits in our studied sections, asthese type of sediments have been reported from

latest Eocene or early Oligocene records in theBlack Sea and South Caspian basins, and havebeen interpreted as indicating the initial isolationand birth of the Paratethys Sea (Baldi 1984; Rusu1985; Dercourt et al. 1993; Robinson et al. 1996;Rogl 1999; Popov et al. 2004, 2008; Schulz et al.2005; Vincent et al. 2005; Allen & Armstrong2008; Johnson et al. 2009). However, the observeddisconformity near the EOT in the SW TarimBasin (Bosboom et al. 2014b) indicates that theCentral Asian basins probably remained hydrologi-cally connected to the western Tethys until the latestEocene and had not yet established their present-dayinternal drainage configurations by basin closure.

Controlling mechanism

We previously discussed the cause of the sea retreatfrom the Tarim Basin, showing that probably a com-bination of long-term regional tectonism and super-imposed short-term global sea-level fluctuationsmay have been responsible for the sea retreat(Bosboom et al. 2014b, c). This study indicates thatthe transgressions and regressions extended across asignificant part of Central Asia, confirming that theunderlying forcing mechanisms must have beenoperating over a large geographical range. Suchmechanisms might have comprised: (1) distal tec-tonism and exhumation of the proto-Kunlun inresponse to the ongoing Indo-Asia collision, lead-ing to widespread increased sedimentation withinthe large extent of the marine basin, which – withvery low relief and transport capacity – was particu-larly susceptible to infilling from sedimentation; (2)eustatic sea-level fall, affecting an extensive areadue to the shallow nature of the proto-ParatethysSea; or (3) a combination of both.

In the Afghan–Tajik Basin, Alai Valley Basinand Fergana Basin, there is more evidence of tec-tonic activity than in the Tarim Basin. The nearlyconstant coarse-grained clastic input in most sec-tions could be an expression of local tectonicdeformation. Our sedimentological analyses in theAfghan–Tajik Basin show that this basin mayhave been particularly affected by tectonic instabil-ity at the top of the fourth marine incursion, as indi-cated by the presence of mica sandstones, reworkingand slumping. These depositional instabilities mayprovide an explanation for the unconformity recog-nized within the fourth incursion, which has not yetbeen identified in the other studied basins of CentralAsia. Hence, the thrust fronts of the western Pamirwere probably more active and less distal comparedto the West Kunlun Shan in the SW Tarim Basin.The mica sandstones are probably of volcanogenicorigin sourced by Cenozoic granitoids in thewestern Central Pamir terrane (Schwab et al.2004). However, studies on the tectonic evolution

R. BOSBOOM ET AL.



of the western side of the Pamir in the Eocene arescarce and, apart from reports of accelerated exhu-mation of the north Central Pamir in the middleEocene (Amidon & Hynek 2010), there is no evi-dence in support of Eocene tectonic activity nearthe Afghan–Tajik Basin.

However, the lack of diachroneity, the discon-nection of the West Siberian Sea and the TurgaiStrait from the Arctic Sea synchronous with thefourth regression in the late Lutetian (Akhmetiev& Beniamovski 2006; Akhmetiev 2007; Iakovleva& Heilmann-Clausen 2010), the late Eocene Euro-pean regressions (e.g. Proust & Hosu 1996; Lopez-Blanco et al. 2000; Lartaud 2007; Costa et al. 2010;Dawber et al. 2011) concomitant with the fifthregression in the late Priabonian, and the observeddisconformity at the EOT in the SW Tarim Basinpalaeodepocentre (Bosboom et al. 2014b, c) indicatethat, superimposed on the long-term tectonicallycontrolled westward trend, short-term eustatic sea-level changes probably forced the individualregressions. Comparison with the global sea-levelcurve of Kominz et al. (2008) shows that the tworegressions studied here could be concomitant withminor drops in global sea level. Hence, these asser-tions further strengthen our previous concept oflong-term westward retreat by early activation ofthe Pamir and superimposed short-term eustaticsea-level fluctuations that forced each individualregression step of the proto-Paratethys Sea fromCentral Asia.

Palaeoenvironmental impact

Previously, we showed that the sea-retreat stepsfrom the westernmost margin of the Tarim Basin(Bosboom et al. 2014b, c) are contemporary withmiddle Eocene East Asian monsoon intensification(Quan et al. 2011, 2012) and aridification stepsrecorded near approximately 41 Ma (base C19n-base C18r) and 37.1 Ma (top C17n.1n) in theXining Basin along the NW margin of the TibetanPlateau (Abels et al. 2011; Bosboom et al. 2014a).This study shows that the early Priabonian fifthregression in the Tarim Basin may have been theonset of the permanent retreat from Central Asia,pushing the eastern margin of the sea from thewesternmost Tarim Basin to the East Aral Basin(Popov et al. 2010), a shift of nearly 1000 km tothe west (Fig. 8). Based on previous climate model-ling results (Ramstein et al. 1997; Zhang et al.2007), it is, indeed, likely that, after this major searetreat, moisture supply to the Asian interior waspermanently and significantly reduced. Zhang et al.(2007) showed that the key criterion for changingthe palaeoenvironmental patterns in China was theretreat to the Turan Plate, which, as we suggesthere, occurred after the final fifth transgression.

The coeval timing of this regression from CentralAsia and the major approximately 37.1 Ma aridifi-cation step is, hence, perfectly in line with theseclimate models and would confirm that the proto-Paratethys Sea had been an important moisturesource for the Asian continental interior up untilthe late Eocene.

Conclusions

New macrofossil data from the Afghan–Tajik andFergana basins and the Alai Valley Basin showthat these basins experienced similar gradual third-order shallow-marine transgressions and regres-sions to the Tarim Basin. Both the fourth transgres-sion of Lutetian age and the fifth transgression oflatest Bartonian–early Priabonian age recognizedin the Tarim Basin can be traced across Central Asia.The fourth transgression comprises two stackedtransgression–regression sequences, and is charac-terized by relatively warm waters and dominatedby carbonate shoals and a platform environment.The marine deposits of the fifth transgression areless widespread and comprise primarily low-energysubtidal and intertidal mudstones. Depositionalinstabilities and continuous clastic input show thatthe basins west of the Tarim Basin may have beenmore tectonically active owing to the early exhuma-tion of the Pamir. After the fifth marine incursioninto Central Asia, the eastern margin of the sea prob-ably shifted hundreds of kilometres westwards,leading to strongly reduced moisture transport tothe Asian interior in the late Eocene.

Our results provide a first-order framework ofthe sea retreat from the major incursion in themiddle Eocene to its final retreat out of CentralAsia in the late Eocene, extending the previousstratigraphic framework that we built up for theTarim Basin further westwards. Potential control-ling mechanisms and environmental impacts of theshort-term sea-level fluctuations and long-termretreat are identified, but further high-resolution agecontrol and detailed sedimentological work arerequired to ascertain the palaeogeographical dyna-mics of the proto-Paratethys Sea in Central Asia.Future magneto-, bio- and sequence-stratigraphicanalyses, in particular, will enable the precise strati-graphic position, age and nature of the palaeogeo-graphical changes identified in the studied basinsto be pinpointed. This work is of utmost importancein making detailed correlations across large dis-tances and, ultimately, offer the required resolu-tion in both space and time to precisely track theevolution of palaeodepocentre migrations in Cen-tral Asia with respect to regional tectonism,palaeoenvironmental change in Asia and global sea-level variation.




The Netherlands Organization for Scientific Research(NWO) and DARIUS funded this project with grants toRoderic Bosboom and Guillaume Dupont-Nivet. Wewould like to thank Laurie Bougeois, Gloria Heilbronn,Jane Qiu, Rajabov Nematjon and Ilhom Oimahmadovfor their contributions in the field. Our thanks also go toFranz Topka (NHM Vienna) for careful preparatorywork on the mollusc material.

Note added in proof

Carrapa et al. 2015, report a Paratethys retreat ca. 39 Ma

based on detrital zircon ages. This corresponds well to

the 4th retreat recorded in the Beshkent formation cor-

related here to the Tarim Basin Wulagen Formation

(Bosboom et al. 2014c). However, the final Paratethys

regression should be attributed to the subsequent 5th

marine transgression-regression reported here in the

Kushan and Shanglak Formations and correlated to the

Tarim Bain Bashibulake Formation. This 5th marine incur-

sion is also shown in the upper part of the WA section of

fig. 2 in Carrapa et al. (2015), several hundred meters

above the ca. 39 Ma detrital zircons. The final retreat is

hence sensibly younger than 39 Ma, assigned here to the

latest Bartonian–early Priabonian.

Appendices

An overview of the applied stratigraphic nomenclature and

the corresponding lithostratigraphic descriptions, thick-

nesses and age estimates from Pomazkov (1972) and

Dzhalilov et al. (1982) are given in Tables A1 & A2,

respectively.

Table A1. Simplified lithostratigraphic description and transgression–regression cyclicity for the Palaeogene stratigraphy of the AlaiValley Basin and Fergana Basin

Alai Valley Basin & Fergana Basin (Kyrgyzstan) Tarim Basin (China)

System Formation Thickness(m)

Age Lithology Formation Sea level

Palaeogene MassagetFm

Oligocene–Miocene

Massive red sandstones andconglomerates

Kezilouyi Fm Continental

ShurysayFm

20–160 Oligocene Brownish red-mudstonesintercalated by siltstones, evaporite

beds and sandstones

BashibulakeFifth Member

Final fifthregression

SumsarFm

0–70 Late Eocene–Oligocene

Reddish-brown mudstones andgrey sandstones rich in oysters and

other bivalves, shark teeth

BashibulakeFourth Member

Final fifthtransgression

HanabadFm

5–70 Late Eocene–Oligocene

Greenish-grey and red (calcareous)mudstones and siltstones, some

marls with bivalves

BashibulakeThird Member

Isfara Fm 5–55 Late Eocene Greenish-grey (calcareous)mudstones and siltstones, somegreen or white sandstones, some

marls with bivalves

BashibulakeSecond Member

RishtanFm

5–60 Late Eocene Red mudstones with thin beds ofred or grey siltstone and sandstone,

some interbeds of marl andlimestone with bivalves

BashibulakeFirst Member

Fourthregression

TurkestanFm

5–150 Middle–lateEocene

Greenish-grey mudstones with greyand white siltstones, sandstones,

marls and limestones, redmudstone intervals at the top, rich

in oysters and other bivalves

Wulagen Fm Fourthtransgression

Alai Fm 10–210 MiddleEocene

Greenish-grey mudstones with greyand white siltstones, marls and

limestones, some gypsiferous redmudstone intervals in middle, rich

in oysters and other bivalves

Kalatar Fm

Suzak Fm 5–120 Early Eocene Complex coloured (calcareous)mudstones, siltstones and

sandstones, bivalves

Upper QimugenFm

Thirdregression

BukharaFm

20–80 Paleocene Limestones, evaporite beds andwhite sandstones with thin

calcareous mudstone layers,bivalves and gastropods

Lower QimugenFm

Thirdtransgression

AkdzharFm

25–125 Paleocene Red mudstones and siltstones withinterbeds of evaporite and

dolomite

Aertashi Fm Secondregression

The corresponding thicknesses and ages of the formations are not to scale and are summarized from Pomazkov (1972). Correlation with thestratigraphic framework of the Tarim Basin is based on results of this study.

R. BOSBOOM ET AL.



Table A2. Simplified lithostratigraphic description and transgression–regression cyclicity for the Palaeogene and early Neogene stratigraphy of

the Afghan–Tajik Basin

Afghan–Tajik Basin (Tajikistan) Tarim Basin (China)

System Formation Thickness

(m)

Age Lithology Formation Sea level

Neogene Baldzhuan

Complex

Kamoli 180–310 Miocene compact massive and coarse

laminated red sandstones

with rare interbeds of

mudstones, siltstones and

conglomerates

Kezilouyi Fm Continental

Palaeogene Shurysay Fm 0–190 Oligocene Reddish-brown mudstones,

sandstones and siltstones

with gypsum, rare interbeds

of limestone

Hissarak Fm 0–130 Oligocene Greenish-grey sandstones and

reddish-brown mudstones,

siltstones and sandstones,

rare gypsum interbeds and

stromatolites at the base

Bashibulake

Fifth Member

Final fifth

regression

Sanglak Fm 0–90 Middle–

late

Eocene

Brownish-red mudstones with

red-brownish-grey sandstones

at the top, bivalves

Bashibulake

Fourth

Member

Final fifth

transgression

Kushan Fm 0–175 Middle–

late

Eocene

Greenish-grey (calcareous)

mudstones with rare

interbeds of limestones and

shell beds, at the top

reddish-brown mudstones

Bashibulake

Third Member

Bashibulake

Second

Member

Tochar Fm Upper 0–175 Middle–

late

Eocene

Red mudstones, siltstones and

sandstones, bivalves

Bashibulake

First Member

Fourth

regression

Lower 0–40 Grey sandstones with mudstone

and white gypsum at the top

Wulagen Fm

Beshkent

Fm

Gandzhin

layers

0–60 Middle

Eocene

Greenish-grey calcareous

mudstones with interbeds of

marls, shell beds, limestones

and sandstones

Fourth

transgression

Turkestan

layers

0–130 Middle

Eocene

Greenish-grey mudstones with

rare shell beds

Jukar Fm 20–240 Middle

Eocene

Greenish-grey mudstones, marls,

limestones, shell beds and

some sandstones, red

mudstones, dolomites and

gypsum interbeds

Kalatar Fm

Givar Fm 5–130 Early

Eocene

Grey mudstones, oysters Upper

Qimugen Fm

Third

regression

Karatag Fm 0–40 Paleocene Grey calcareous mudstones,

marls and muddy limestones

Lower

Qimugen Fm

Third

transgression

Arukfan Fm 0–60 Paleocene Grey calcareous mudstones,

marls and limestones,

gypsum beds and dolomite at

the top, molluscs

Tabakcha Fm 5–130 Paleocene Limestones with rare beds of

dolomite and gypsum,

bivalves

Akdjar Fm 10–250 Paleocene Gypsum, dolomites and red

mudstones, some interbeds of

red siltstones and sandstones,

bivalves

Aertashi Fm Second

regression

The corresponding thicknesses and ages of the formations are not to scale and summarized from Dzhalilov et al. (1982). Correlation withthe stratigraphic framework of the Tarim Basin is based on results of this study.




References

Abels, H. A., Dupont-Nivet, G., Xiao, G., Bosboom,R. E. & Krijgsman, W. 2011. Step-wise Asianpaleoenvironmental changes preceding the Eocene–Oligocene transition (EOT) in the terrestrial XiningBasin, China. Palaeogeography, Palaeoclimatology,Palaeoecology, 299, 399–412.

Aigner, T., Brandenburg, A., van Vliet, A., Lawr-

ence, D. & Westrich, J. 1990. Stratigraphic model-ling of epicontinental basins: two applications.Sedimentary Geology, 69, 167–190.

Akhmetiev, M. A. 2007. Paleocene and eocene floras ofRussia and adjacent regions: climatic conditions oftheir development. Paleontological Journal, 41,1032–1039.

Akhmetiev, M. A. & Beniamovski, V. N. 2006. ThePaleocene and Eocene in the Russian part of WestEurasia. Stratigraphy and Geological Correlation,14, 49–72.

Allen, M. B. & Armstrong, H. A. 2008. Arabia–Eurasiacollision and the forcing of Mid-Cenozoic global cool-ing. Palaeogeography, Palaeoclimatology, Palaeoe-cology, 265, 52–58.

Allen, M. B. & Armstrong, H. A. 2012. Reconcilingthe Intertropical Convergence Zone, Himalayan/Tibetan tectonics, and the onset of the Asian mon-soon system. Journal of Asian Earth Sciences, 44,36–47.

Amidon, W. H. & Hynek, S. A. 2010. Exhumationalhistory of the North Central Pamir. Tectonics, 29,TC5017.

An, Z., Kutzbach, J. E., Prell, W. L. & Porter, S. C.2001. Evolution of Asian monsoons and phased upliftof the Himalaya–Tibetan Plateau since Late Miocenetimes. Nature, 411, 62–66.

Baldi, T. 1973. Mollusc Fauna of the Hungarian UpperOligocene (Egerian). Studies in Stratigraphy, Palaeoe-cology, Palaeogeography and Systematics. AkademiaKiado, Budapest.

Baldi, T. 1984. The Terminal Eocene and Early OligoceneEvents in Hungary and the Separation of an Anoxic,Cold Paratethys. Eclogae Geologicae Helvetiae, 77,1–27.

Barker, P. F., Diekmann, B. & Escutia, C. 2007. Onsetof Cenozoic Antarctic glaciation. Deep Sea ResearchPart II: Topical Studies in Oceanography, 54,2293–2307.

Bazhenov, M. L. & Burtman, V. S. 1981. Formation ofthe Pamir–Penjab syntaxis: implications from paleo-magnetic investigations of Lower Cretaceous andPaleogene rocks of the Pamirs. In: Sinha, A. K. (ed.)Contemporary Geoscientific Researches in Himalaya,Volume 1. A Commemorative Volume in Honour ofS. P. Nautiyal. Bishen Singh Mahendra Pal Sing,Dehra Dun, India, 71–81.

Bazhenov, M. L., Burtman, V. S. & Gurariy, G. Z.1978. Study of the curvature of the Pamirs Arc inPaleogene by paleornagnetic method. Doklady Akade-mii Nauk SSR, 242, 1137–1139.

Berizzi Quarto di Palo, A. 1970. Paleogene Pelecypodsfrom Kataghan and Badakhshan (North-East Afgani-stan). In: Desio, A. (ed.) Fossils of North-East Afgani-stan. Italian Expeditions to the Karakorum (K2), and

Hindu Kush, Scientific Reports I–V. Brill, Leiden,161–240.

Bershaw, J., Garzione, C. N., Schoenbohm, L.,Gehrels, G. & Tao, L. 2012. Cenozoic evolution ofthe Pamir Plateau based on stratigraphy, zircon prove-nance, and stable isotopes of foreland basinsediments at Oytag (Wuyitake) in the Tarim Basin(West China). Journal of Asian Earth Sciences, 44,136–148.

Bohaty, S. M. & Zachos, J. C. 2003. Significant SouthernOcean warming event in the Late Middle Eocene.Geology, 31, 1017–1020.

Bohaty, S. M., Zachos, J. C., Florindo, F. & Delaney,M. L. 2009. Coupled greenhouse warming and deepsea acidification in the Middle Eocene. Paleoceano-graphy, 24, 1–16.

Bosboom, R. E., Dupont-Nivet, G. et al. 2011. LateEocene sea retreat from the Tarim Basin (WestChina) and concomitant asian paleoenvironmentalchange. Palaeogeography, Palaeoclimatology, Palaeoe-cology, 299, 385–398.

Bosboom, R. E., Abels, H. A., Hoorn, C., van den Berg,B. C. J., Guo, Z. & Dupont-Nivet, G. 2014a.Descent into icehouse-climate state on the Asian con-tinent after the Middle Eocene Climatic Optimum(MECO). Earth and Planetary Science Letters, 389,34–42.

Bosboom, R. E., Dupont-Nivet, G. et al. 2014b.Linking Tarim Basin sea retreat (west China) andAsian aridification in the late Eocene. Basin Research,26, 621–640.

Bosboom, R. E., Dupont-Nivet, G. et al. 2014c. Tim-ing, cause and impact of the Late Eocene stepwisesea retreat from the Tarim Basin (west China). Palaeo-geography, Palaeoclimatology, Palaeoecology, 403,101–118.

Brookfield, M. E. & Hashmat, A. 2001. The geologyand petroleum potential of the North Afghan platformand adjacent areas (northern Afghanistan, with parts ofsouthern Turkmenistan, Uzbekistan and Tajikistan).Earth Science Reviews, 55, 41–71.

Budd, D. A. & Harris, P. M. (eds). 1990. Carbonate–Siliciclastic Mixtures. SEPM Reprint Series, 14.SEPM (Society for Sedimentary Geology), Tulsa, OK.

Burtman, V. S. 2000. Cenozoic crustal shorteningbetween the Pamir and Tien Shan and a reconstruc-tion of the Pamir–Tien Shan transition zone for theCretaceous and Palaeogene. Tectonophysics, 319,69–92.

Burtman, V. S. & Molnar, P. 1993. Geological andGeophysical Evidence for Deep Subduction of Conti-nental Crust Beneath the Pamir. Geological Societyof America, Special Papers, 281.

Burtman, V. S., Skobelev, S. F. & Molnar, P. 1996.Late Cenozoic slip on the Talas–Ferghana Fault, theTien Shan, Central Asia. Geological Society ofAmerica Bulletin, 108, 1004–1021.

Carrapa, B., Decelles, P. G. et al. 2015. Tectono-cli-matic implications of Eocene Paratethys regression inthe Tajik basin of central Asia. Earth and PlanetaryScience Letters, 424, 168–178.

Conquand, H. 1862. Geologie et paleontologie de laregion sud de la province de Constantine. Memoiresde la Societe d’Emulation de Provence, 2.

R. BOSBOOM ET AL.



Costa, E., Garces, M., Lopez Blanco, M., Beamud, E.,Gomez-Paccard, M. & Larrasoana, J. C. 2010.Closing and continentalization of the South PyreneanForeland Basin (NE Spain): Magnetochronologicalconstraints. Basin Research, 22, 904–917.

Coutand, I., Strecker, M. R., Arrowsmith, R., Hilley,G., Thiede, R. C., Korjenkov, A. & Omuraliev, M.2002. Late Cenozoic tectonic development of the intra-montane Alai Valley, (Pamir–Tien Shan region,Central Asia): an example of intracontinental defor-mation due to the Indo-Eurasia collision. Tectonics,21, 3-1–3-19.

Cowgill, E. 2010. Cenozoic right-slip faulting along theeastern margin of the Pamir salient, northwesternChina. Geological Society of America Bulletin, 122,145–161.

Cox, L. R. 1936. Fossil Mollusca from Southern Persia(Iran) and Bahrein Island. Geological Survey ofIndia, Memoirs, Palaeontologia Indica (NS), 22.

Cox, L. R. 1938. Fossiles Eocenes du nord de l’Afghani-stan. Memoires de la Societe Geologique de France,39, 29–44.

Cox, L. R. 1962. Tertiary Bivalvia from Libya. Palaeon-tology, 5, 1–8.

Davidzon, R. M., Kreydenkov, G. P. & Salibaev, G. H.1982. Stratigraphy of Palaeogene Deposits of theTajik Depression and Adjacent Areas. Donish,Dushanbe.

Dawber, C. F., Tripati, A. K., Gale, A. S., MacNio-

caill, C. & Hesselbo, S. P. 2011. Glacioeustasyduring the Middle Eocene? Insights from the stratigra-phy of the Hampshire Basin, UK. Palaeogeography,Palaeoclimatology, Palaeoecology, 300, 84–100.

De Grave, J., Glorie, S. et al. 2012. Late palaeozoic andMeso-Cenozoic tectonic evolution of the SouthernKyrgyz Tien Shan: constraints from multi-methodthermochronology in the Trans-Alai, Turkestan–Alaisection and the southeastern Ferghana Basin. Journalof Asian Earth Sciences, 44, 149–168.

Dercourt, J., Ricou, L. E. & Vrielynck, B. (eds). 1993.Atlas Tethys Paleoenvironmental Maps. Commisionfor the Geological Map of the World, Paris.

Dupont-Nivet, G., Krijgsman, W., Langereis, C. G.,Abels, H. A., Dai, S. & Fang, X. 2007. Tibetanplateau aridification linked to global cooling at theEocene–Oligocene transition. Nature, 445, 635–638.

Dupont-Nivet, G., Hoorn, C. & Konert, M. 2008.Tibetan uplift prior to the Eocene–Oligocene Cli-mate transition: evidence from pollen analysis of theXining Basin. Geology, 36, 987–990.

Dzhalilov, M. R., Alekseev, M. N., Andreev, Y. N. &Salibaev, G. K. 1982. Mesozoic and Cenozoicdeposits of the northern part of the Afghano-TajikBasin. In: Stratigraphic Correlation Between Sedimen-tary Basins of the ESCAP Region, Volume 8. ESCAPAtlas of Stratigraphy 3 – Australia, Bangladesh, Fiji,India, Indonesia, Nepal, Solomon Islands, Tajikistan.Mineral Resources Development Series, 48. UnitedNations (Economic and Social Commission for Asiaand the Pacific), New York, 24–32.

Edgar, K. M., Wilson, P. A., Sexton, P. F. & Suga-

numa, Y. 2007. No extreme bipolar glaciation duringthe main Eocene calcite compensation shift. Nature,448, 908–911.

Edgar, K. M., Wilson, P. A., Sexton, P. F., Gibbs, S. J.,Roberts, A. P. & Norris, R. D. 2010. New biostrati-graphic, magnetostratigraphic and isotopic insightsinto the Middle Eocene Climatic Optimum in low lati-tudes. Palaeogeography, Palaeoclimatology, Palaeoe-cology, 297, 670–682.

Gasson, E., Siddall, M., Lunt, D. J., Rackham, O. J. L.,Lear, C. H. & Pollard, D. 2012. Exploring uncertain-ties in the relationship between temperature, icevolume, and sea level over the past 50 million years.Reviews of Geophysics, 50, 1–35.

Gomes-Rodrigues, H., Marivaux, L. & Vianey-Liaud,M. 2012. Expansion of open landscapes in NorthernChina during the Oligocene induced by dramaticclimate changes: paleoecological evidence. Palaeo-geography, Palaeoclimatology, Palaeoecology, 358–360, 62–71.

Gradstein, F. M., Ogg, J. G., Schmitz, M. & Ogg, G.2012. The Geologic Time Scale 2012. Cambridge Uni-versity Press, Cambridge.

Graham, S. A., Chamberlain, C. P. et al. 2005. Stableisotope records of Cenozoic climate and topography,Tibetan Plateau and Tarim Basin. American Journalof Science, 305, 101–118.

Grewingk, C. 1853. Die Geognostischen Und Orogra-phischen Verhaeltnisse Des Noerdlichen Persiens. Ver-handlungen der R.K Mineralogischen Gesellschaft,1852–1853, 97–245.

Hao, Y. C. & Zeng, X. L. 1984. On the evolution of theWest Tarim Gulf from Mesozoic to Cenozoic interms of characteristics of foraminiferal Fauna. ActaMicropalaeontologica Sinica, 1, 1–13.

Hendrix, M. S., Graham, S. A., Carroll, A. R.,Sobel, E. R., Mcknight, C. L., Schulein, B. J. &Wang, Z. 1992. Sedimentary record and climaticimplications of recurrent deformation in the TianShan: evidence from Mesozoic Strata of the NorthTarim, South Junggar, and Turpan Basins, NorthwestChina. Geological Society of America Bulletin, 104,53–79.

Iakovleva, A. I. & Heilmann-Clausen, C. 2010.Eocene dinoflagellate cyst biostratigraphy of researchborehole 011-BP, Omsk region, southwestern Siberia.Palynology, 34, 195–232.

Ilyina, A. P. 1955. Paleogene Molluscs from the NorthernUstyurt. Trudy VNIGRI, 89.

Jarvis, A., Reuter, H. I., Nelson, A. & Guevara, E.2008. Hole-Filled Seamless Srtm Data V4. Inter-national Centre for Tropical Agriculture (CIAT), avail-able from http://srtm.csi.cgiar.org.

Johnson, C. L., Hudson, S. M., Rowe, H. D. &Efendiyeva, M. A. 2009. Geochemical constraintson the Palaeocene–Miocene evolution of easternAzerbaijan, with implications for the South CaspianBasin and eastern Paratethys. Basin Research, 22,733–750.

Jolivet, M., Brunel, M. et al. 2001. Mesozoic andCenozoic tectonics of the northern edge of theTibetan Plateau: fission-track constraints. Tectonophy-sics, 343, 111–134.

Karagiuleva, J. D. 1964. Les Fossiles De Bulgarie.Academy of Science Bulgaria, Sofia.

Katz, M. E., Miller, K. G., Wright, J. D., Wade, B. S.,Browning, J. V., Cramer, B. S. & Rosenthal, Y.



http://srtm.csi.cgiar.org

http://srtm.csi.cgiar.org


2008. Stepwise transition from the Eocene greenhouseto the Oligocene icehouse. Nature Geoscience, 1,329–334.

Klootwijk, C. T. 1979. A summary of paleomagnetic datafrom extra-peninsular Indo-Pakistan and South CentralAsia: implications for collision tectonics. In: Saklani,P. S. (ed.) Structural Geology of the Himalaya. Today& Tomorrow Printers & Publishers, New Delhi,307–360.

Kominz, M. A., Browning, J. V., Miller, K. G., Sugar-

man, P. J., Mizintseva, S. & Scotese, C. R. 2008.Late Cretaceous to Miocene sea-level estimates fromthe New Jersey and Delaware coastal plain coreholes:An error analysis. Basin Research, 20, 211–226.

Kraatz, B. P. & Geisler, J. H. 2010. Eocene–Oligocenetransition in Central Asia and its effects on mammalianevolution. Geology, 38, 111–114.

Krasheninnikov, V. A. & Akhmetiev, M. A. (eds).1998. Late Eocene–Early Oligocene Geological andBiotical Events on the Territory of the Former SovietUnion. Part II: The Geological and Biotical Events.Geos, Moscow.

Lan, X. 1997. Paleogene bivalve communities in theWestern Tarim Basin and their paleoenvironmentalimplications. Paleoworld, 7, 137–157.

Lan, X. & Wei, J. (eds). 1995. Late Cretaceous–EarlyTertiary Marine Bivalve Fauna from the WesternTarim Basin. Chinese Science Publishing House,Beijing.

Lartaud, F. 2007. Les Fluctuation Haute Frequence del’environnement au Cours des Temps Geologiques.Mise au Point d’un Modele de Refeference Actuelsur l’enregistrement des Contrastes Saisonniers dansl’Atlantique Nord. Universite Pierre et Marie Curie,Paris.

Lear, C. H., Bailey, T. R., Pearson, P. N., Coxall, H. K.& Rosenthal, Y. 2008. Cooling and ice growth acrossthe Eocene–Oligocene transition. Geology, 36,251–254.

Lopez-Blanco, M., Marzo, M., Burbank, D. W.,Verges, J., Roca, E., Anadon, P. & Pina, J. 2000.Tectonic and climatic controls on the development offoreland fan deltas: Montserrat and Sant LlorencDel Munt systems (Middle Eocene, Ebro Basin, NESpain). Sedimentary Geology, 138, 17–39.

Lyle, M., Olivarez-Lyle, A., Backman, J. & Tripati, A.2005. Biogenic sedimentation in the Eocene Equato-rial Pacific – the stuttering greenhouse and Eocene car-bonate compensation depth. In: Wilson, P. A., Lyle,M. & Firth, J. V. (eds) Proceedings of the Ocean Dril-ling Program, Scientific Results, 199. Ocean DrillingProgram, College Station, TX, 1–35.

Mao, S. & Norris, G. 1988. Late Cretaceous–Early Ter-tiary Dinoflagellates and Acritarchs from the KashiArea, Tarim Basin, Xinjiang Province, China. RoyalOntario Museum, Toronto.

Markowsky, A. (ed.). 1959. Tajik SSR. USSR GeologicalDescription. State of Scientific and Technical Publish-ing, Moscow.

Meng, Z., Deng, Y., Ding, Z., Li, Y. & Sun, D. 1998.New paleomagnetic results from Ceno-Mesozoic vol-canic rocks along southern rim of the Tarim Basin,China. Science in China Series D Earth Sciences(English edition), 41, 91–104.

Miller, K. G., Kominz, M. A. et al. 2005. The Phaner-ozoic record of global sea-level change. Science, 310,1293–1298.

Molnar, P. & Tapponnier, P. 1975. Cenozoic tectonicsof Asia: effects of a continental collision. Science,189, 419–426.

Nikolaev, V. G. 2002. Afghan–Tajik depression: archi-tecture of sedimentary cover and evolution. RussianJournal of Earth Sciences, 4, 399–421.

Ovechkin, N. K. 1954. The Middle Paleogene Deposits ofthe Turgai Passage and the Northern Aral Region.Trudy VSEGEI, Moscow.

Pekar, S. F., Christie-Blick, N., Kominz, M. A. &Miller, K. G. 2002. Calibration between eustatic esti-mates from backstripping and oxygen isotopic recordsfor the Oligocene. Geology, 30, 903.

Pomazkov, K. D. (ed.). 1972. Kyrgyz Ssr Geolog-ical Description. USSR Geology Publishing House(NEWPA), Moscow.

Popov, S., Rogl, F., Rozanov, A. Y., Steininger, F. F.,Shcherba, I. G. & Kovac, M. 2004. Lithological–Paleogeographic Maps of Paratethys 10 Maps LateEocene to Pliocene. Courier Forschungsinstitut Senck-enberg, 250.

Popov, S. V., Sychevskaya, E. K., Akhmetiev, M. A.,Zaporozhets, N. I. & Golovina, L. A. 2008. Strati-graphy of the maikop group and pteropoda beds inNorthern Azerbaijan. Stratigraphy and GeologicalCorrelation, 16, 664–677.

Popov, S. V., Antipov, M. P., Zastrozhnov, A. S.,Kurina, E. E. & Pinchuk, T. N. 2010. Sea-level fluc-tuations on the northern shelf of the eastern Paratethysin the Oligocene–Neogene. Stratigraphy and Geologi-cal Correlation, 18, 200–224.

Proust, J. N. & Hosu, A. 1996. Sequence stratigraphyand paleogene tectonic evolution of the Transylvan-ian Basin (Romania, Eastern Europe). SedimentaryGeology, 105, 117–140.

Quan, C., Liu, Y. S. C. & Utescher, T. 2011. Paleogeneevolution of precipitation in northeastern China sup-porting the Middle Eocene intensification of the EastAsian monsoon. PALAIOS, 26, 743–753.

Quan, C., Liu, Y.-S. C. & Utescher, T. 2012. Paleo-gene temperature gradient, seasonal variation andclimate evolution of northeast China. Palaeogeogra-phy, Palaeoclimatology. Palaeoecology, 313–314,150–161.

Ramstein, G., Fluteau, F., Besse, J. & Joussaume, S.1997. Effect of orogeny, plate motion and land–seadistribution on Eurasian climate change over the past30 million years. Nature, 386, 788–795.

Raulin, V. & Delbos, J. 1855. Extrait d’une monographiedes Ostrea des terrains tertiaires de l’Aquitaine. Bulle-tin de la Societe Geologique de France, serie 2, 12,1144–1164.

Reading, H. G. (ed.). 2006. Sedimentary Environ-ments: Processes, Facies and Stratigraphy. Blackwell,Oxford.

Robinson, A. G., Rudat, J. H., Banks, C. J. & Wiles,R. L. F. 1996. Petroleum geology of the Black Sea.Marine and Petroleum Geology, 13, 195–223.

Robinson, D. M., Dupont-Nivet, G., Gehrels, G. E. &Zhang, Y. 2003. The Tula uplift, northwestern China;evidence for regional tectonism of the northern Tibetan

R. BOSBOOM ET AL.



Plateau during late Mesozoic–early Cenozoic time.Geological Society of America Bulletin, 115, 35–47.

Rogl, F. 1999. Short Note: Mediterranean and Paratethys.Facts and hypotheses of an Oligocene to Miocenepaleogeography (short overview). Geologica Car-pathica, 50, 339–349.

Romanovskiy, G. D. 1879–1890. Materials for theGeology of the Turkestan region. Academie Imperialedes Sciences, St. Petersburg. Volume 1 (1879),Volume 2 (1884), Volume 3 (1890).

Rusu, A. 1985. Oligocene events in Transylvania (Rouma-nia) and the first separation of paratethys. Dari deSeama ale Sedintelor Institutul de Geologie si Geofi-zica, 72–73, 207–223.

Rusu, A., Brotea, D. & Melinte, M. C. 2004. Biostrati-graphy of the Bartonian deposits from Gilau area (NWTransylvania, Romania). Acta Palaeontologica Roma-niae, 4, 441–454.

Salibaev, G. H. 1972. Molluski Verhnei Chasti Hana-badskih I Shumsharskih Sloev Tadzhikskoi Depressii.Akademiji fanhoi RSS Tochikiston, Dushanbe.

Schulz, H.-M., Bechtel, A. & Sachsenhofer, R. F.2005. The birth of the Paratethys during the Early Oli-gocene: from Tethys to an ancient Black Sea analogue?Global and Planetary Change, 49, 163–176.

Schwab, M., Ratschbacher, L. et al. 2004. Assembly ofthe Pamirs: age and origin of magmatic belts from thesouthern Tien Shan to the southern Pamirs and theirrelation to Tibet. Tectonics, 23, TC4002.

Sharpe, D. 1850. On the secondary district of Portugalwhich lies on the north of the Tagus. Quarterly Journalof the Geological Society, London, 6, 135–195,http://doi.org/10.1144/GSL.JGS.1850.006.01-02.18

Sobel, E. R. & Dumitru, T. A. 1997. Thrusting and exhu-mation around the margins of the western Tarim Basinduring the India–Asia collision. Journal of Geophysi-cal Research, 102, 5043–5063.

Sobel, E. R., Chen, J. & Heermance, R. V. 2006. LateOligocene–Early Miocene initiation of shortening inthe Southwestern Chinese Tian Shan: implicationsfor Neogene shortening rate variations. Earth and Pla-netary Science Letters, 247, 70–81.

Sobel, E. R., Chen, J., Schoenbohm, L. M., Thiede, R.,Stockli, D. F., Sudo, M. & Strecker, M. R. 2013.Oceanic-style subduction controls late Cenozoic defor-mation of the northern Pamir orogen. Earth and Plane-tary Science Letters, 363, 204–218.

Solander, D. C. 1766. Fossilia Hantoniensia Collecta, Etin Musaeo Britannico Deposita, a Gustavo Brander.Gustavus Brander, London.

Stenzel, H. B. 1971. Oysters. In: Moore, R. C. (ed.)Treatise on Invertebrate Paleontology, Volume 3. Geo-logical Society of America/University of KansasPress, Boulder, CO/Lawrence, KS, 953–1224.

Strecker, M. R., Hilley, G. E., Arrowsmith, J. R. &Coutand, I. 2003. Differential structural and geo-morphic mountain-front evolution in an active conti-nental collision zone: the northwest Pamir, southernKyrgyzstan. Bulletin of the Geological Society ofAmerica, 115, 166–181.

Sun, X. & Wang, P. 2005. How old is the Asian monsoonsystem? Palaeobotanical records from China. Palaeo-geography, Palaeoclimatology, Palaeoecology, 222,181–222.

Tang, T., Yang, H. et al. 1989. Marine Late Creta-ceous and Early Tertiary Stratigraphy and PetroleumGeology in Western Tarim Basin, China. SciencePress, Beijing.

Tang, T., Xue, Y. & Yu, C. 1992. Characteristics andSedimentary Environments of the Late Cretaceous toEarly Tertiary Marine Strata in the Western Tarim,China. Science Press, Beijing.

Thomas, J.-C., Perroud, H., Cobbold, P. R., Bazhenov,M. L., Burtman, V. S., Chauvin, A. & Sadybaka-

sov, I. S. 1993. A paleomagnetic study of Tertiary for-mations from the Kyrgyz Tien-Shan and its tectonicimplications. Journal of Geophysical Research, 98,9571–9589.

Thomas, J.-C., Chauvin, A., Gapais, D., Bazhenov,M. L., Perroud, H., Cobbold, P. R. & Burtman,V. S. 1994. Paleomagnetic evidence for Cenozoicblock rotations in the Tadjik depression (Central Asia).Journal of Geophysical Research, 99, 15,141–15,160.

Tripathy-Lang, A., Hodges, K. V., van Soest, M. C. &Ahmad, T. 2013. Evidence of pre-Oligocene emer-gence of the Indian passive margin and the timing ofcollision initiation between India and Eurasia. Litho-sphere, 5, 501–506.

Tripati, A., Backman, J., Elderfield, H. & Ferretti, P.2005. Eocene bipolar glaciation associated with globalcarbon cycle changes. Nature, 436, 341–346.

Tripati, A. K., Eagle, R. A. et al. 2008. Evi-dence for glaciation in the northern hemisphere backto 44 Ma from ice-rafted debris in the Green-land Sea. Earth and Planetary Science Letters, 265,112–122.

van Hinsbergen, D. J. J., Lippert, P. C., Dupont-Nivet,G., McQuarrie, N., Doubrovine, P. V., Spakman,W. & Torsvik, T. H. 2012. Greater India Basin hypoth-esis and a two-stage Cenozoic Collision between Indiaand Asia. Proceedings of the National Academy ofSciences, 109, 7659–7664.

Vialov, O. S. 1937. Sur La Clasification Des Ostreides EtLeur Valeur Stratigraphique. 128 Congres Inter-national de Zoologie, Comptes Rendus, 3, 1627–1638.

Vialov, O. S. 1938. On the Middle-Asiatic Fatina Vialovand Turkostrea Vialov. Izvestija Akademii NaukSSSR, Moscow.

Vialov, O. S. 1948. Paleogenovie Ustrici TadzhikskoiDepressii (Paleogene Ostreids from Tajik Depression).Trudy VNIGRI, Leningrad.

Villa, G., Fioroni, F., Pea, L., Bohaty, S. & Persico, D.2008. Middle Eocene–late Oligocene climate vari-ability: calcareous nannofossil response at KerguelenPlateau, Site 748. Marine Micropaleontology, 69,173–192.

Vincent, S. J., Allen, M. B., Ismail-Zadeh, A. D.,Flecker, R., Foland, K. A. & Simmons, M. D.2005. Insights from the Talysh of Azerbaijan into thePaleogene evolution of the South Caspian Region.Geological Society of America Bulletin, 117,1513–1533.

Yang, H., Jiang, X. & Lin, S. 1995. Late Cretaceous–Early Tertiary Ostracod Fauna from Western TarimBasin, South Xinjiang, China. Science Press, Beijing.

Yang, W., Jolivet, M., Dupont-Nivet, G. & Guo, Z.2014. Mesozoic–Cenozoic tectonic evolution of south-western Tian Shan: Evidence from detrital zircon



http://doi.org/10.1144/GSL.JGS.1850.006.01-02.18




U/Pb and apatite fission track ages of the Ulugqat area,Northwest China. Gondwana Research, 26, 986–1008,http://doi.org/10.1016/j.gr.2013.07.020

Yang, Y. & Liu, M. 2002. Cenozoic deformation of theTarim Plate and the implications for mountain buildingin the Tibetan Plateau and the Tian Shan. Tectonics,21, 1059.

Yin, A. & Harrison, M. T. 2000. Geologic evolution ofthe Himalayan–Tibetan orogen. Annual Review ofEarth and Planetary Sciences, 28, 211–280.

Yin, A., Rumelhart, P. E. et al. 2002. Tectonic historyof the Altyn Tagh Fault System in Northern TibetInferred from Cenozoic Sedimentation. GeologicalSociety of America Bulletin, 114, 1257–1295.

Zachos, J. C., Dickens, G. R. & Zeebe, R. E. 2008. AnEarly Cenozoic perspective on greenhouse warmingand carbon-cycle dynamics. Nature, 451, 279–283.

Zhang, Z., Wang, H., Guo, Z. & Jiang, D. 2007. Whattriggers the transition of palaeoenvironmental patternsin China, the Tibetan Plateau uplift or the Paratethyssea retreat? Palaeogeography, Palaeoclimatology,Palaeoecology, 245, 317–331.

Zhang, Z., Flatøy, F., Wang, H., Bethke, I.,Bentsen, M. & Guo, Z. 2012. Early Eocene Asianclimate dominated by desert and steppe with limitedmonsoons. Journal of Asian Earth Sciences, 44,24–35.

Zhong, S. 1992. Calcareous Nannofossils from the UpperCretaceous and Lower Tertiary in the Western TarimBasin, South Xinjiang, China. Chinese Science Pub-lishing House, Beijing.

Zubovich, A. V., Wang, X. Q. et al. 2010. GPS velocityfield for the Tien Shan and surrounding regions. Tec-tonics, 29, TC6014.

R. BOSBOOM ET AL.


http://doi.org/10.1016/j.gr.2013.07.020

http://doi.org/10.1016/j.gr.2013.07.020


Documents

Late Eocene palaeogeography of the proto-Paratethys Sea in ...forth/publications/Bosboom_2015.pdf · Late Eocene palaeogeography of the proto-Paratethys Sea in Central Asia (NW China,