Sdarticle

ELSEVIER

Sedimentary Geology 123 (1999) 31–62

Sedimentology of the Cretaceous Maha Sarakham evaporites in the Khorat Plateau of northeastern Thailand Mohamed El Tabakh a,Ł , Cherdsak Utha-Aroon b , B. Charlotte Schreiber c a

School of Applied Geology, Curtin University, G.P.O. Box U 1987, Perth, WA 6001, Australia Geology Division, Department of Mineral Resources, Rama IV, Bangkok, Thailand c Department of Geology, Appalachian State University, 118 Rankin, Science Building, Boone, NC 28608, USA

b Economic

Received 14 May 1997; accepted 8 June 1998

Abstract Evaporites of the Cretaceous to early Tertiary Maha Sarakham Formation on the Khorat Plateau of southeast Asia (Thailand and Laos) are composed of three depositional members that each include evaporitic successions, each overlain by non-marine clastic red beds, and are present in both the Khorat and the Sakon Nakhon sub-basins. These two basins are presently separated by the northwest-trending Phu Phan anticline. The thickness of the formation averages 250 m but is up to 1.1 km thick in some areas. In both basins it thickens towards the basin centre suggesting differential basin subsidence preceding or during sedimentation. The stratigraphy, lithological character and mineralogy of the evaporites and clastics are identical in both basins suggesting that they were probably connected during deposition. Evaporites include thick successions of halite, anhydrite and a considerable accumulation of potassic minerals (sylvite and carnallite) but contain some tachyhydrite, and minor amounts of borates. During the deposition of halite the basin was subjected to repeated inflow of fresher marine water that resulted in the formation of anhydrite marker beds. Sedimentary facies and textures of both halite and anhydrite suggest deposition in a shallow saline-pan environment. Many halite beds, however, contain a curious ‘sieve-like’ fabric marked by skeletal anhydrite outlines of gypsum precursor crystals and are the product of early diagenetic replacement by halite of primary shallow-water gypsum. The δ34 S isotopic values obtained from different types of anhydrite interbedded with halite range from 14.3‰ to 17.0‰ (CDT), suggesting a marine origin for this sulphate. Bromine concentration in the halite of the Lower Member begins around 70 ppm and systematically increases upward to 400 ppm below the potash-rich zone, also suggesting evaporation of largely marine waters. In the Middle Member the initial concentration of bromine in halite is 200 ppm, rising to 450 ppm in the upper part of this member. The bromine concentration in the Upper Member exhibits uniform upward increase and ranges from 200 to 300 ppm. The presence of tachyhydrite in association with the potassic salts was probably the result of: (1) the large volumes of halite replacement of gypsum, on a bed by bed basis, releasing calcium back into the restricted waters of the basin; and (2) early hydrothermal input of calcium chloride-rich waters. The borates associated with potash-rich beds likely resulted from erosion and influx of water from surrounding granitic terrains; however, hydrothermal influx is also possible. Interbedded with the evaporites are non-marine red beds that are also evaporative, with displacive anhydrite nodules and beds and considerable amounts of displacive halite. The δ34 S isotopic values of this anhydrite have non-marine values, ranging from 6.4‰ to 10.9‰ (CDT).

Ł Corresponding

author. E-mail: [email protected]

c 1999 Elsevier Science B.V. All rights reserved. 0037-0738/99/$ – see front matter PII: S 0 0 3 7 - 0 7 3 8 ( 9 8 ) 0 0 0 8 3 - 9

32

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

These data indicate that the Khorat and Sakhon Nakhon basins underwent periods of marine influx due to relative world sea-level rise but were sporadically isolated from the world ocean.  1999 Elsevier Science B.V. All rights reserved. Keywords: Khorat Plateau; Maha Sarakham; potash; evaporites

1. Introduction A long recognised problem in evaporite deposition is the origin and development of saline giants found in the rock record (Sloss, 1969; Hsu¨, 1972; Schreiber, 1988; Busson and Schreiber, 1997). The Maha Sarakham Formation (Cretaceous through Tertiary) on the Khorat Plateau of northeastern Thailand is a saline giant that includes one of the largest salt deposits in the world. The formation is composed of three depositional successions present in the northern Sakon Nakhon and the southern Khorat basins (Fig. 1). Lithologies of the formation are dominated by halite with potassic minerals of sylvite (KCl) and carnallite (KCl MgCl2 .6H2 O), red beds and minor anhydrite, tachyhydrite (CaCl2 .MgCl2 .12H2 O) and borates [hilgardite, Ca2 BCl (OH)2 and boracite, Mg3 ClB7 O13 ]. The origin of these potash minerals is controversial because there are no specific internal features that define depositional or diagenetic features and no modern potash-forming basin of marine origin is active in recent time, hence we have no observable analogue. Preserved petrographic textures of evaporites of halite and gypsum and the paragenesis of the carnallite=sylvite assemblage in the Maha Sarakham Formation provide valuable information about its environment of deposition and early diagenesis. Salt beds of the Maha Sarakham Formation were first discovered in groundwater wells in the Khorat Plateau (La Moreaux et al., 1959), and a preliminary investigation (Gardner et al., 1967) was carried out. Later, sylvite was found near Vientiane, the capital city of Laos, and shortly thereafter, carnallite as well as sylvite were found on the Thai side. Earlier works by Hite (1971, 1974) and Hite and Japakasetr (1979) have outlined the general stratigraphy of the salt sequence and a broad overview of the depositional elements of the Maha Sarakham Formation. Hite and Japakasetr (1979) reported a sharp boundary between the lower part of the evaporite section and the underlying Khok Kruat Formation, suggesting a pos-

sible disconformity. They also present details of the bromine chemistry of the salt deposits and propose a stratigraphic, environmental and structural evolution of the salt beds. Utha-Aroon (1993) suggested nonmarine depositional environment of the red-coloured clastics interbedded with the rock salt deposits based on their sedimentary features. Most giant evaporite deposits are associated with marine shelf-carbonate sequences. The evaporites of the Maha Sarakham Formation, however, lie atop a thick non-marine sequence of the Mesozoic Khorat Group, are interbedded with non-marine red beds, apparently lack the more usual carbonates, and are found in an inland basin on continental crust. All of these features suggest a non-marine origin for the Maha Sarakham salts. The objectives of this paper are to combine details of the depositional features, petrography, and geochemistry of the Maha Sarakham evaporites in order to more define the origin of these evaporites and their diagenetic development and to document those features that are related to other similar saline deposits in the world.

2. Material and methods This study is based on an examination of 235 drilled cores obtained and archived by the D.M.R. (Department of Mineral Resources) of Thailand (Fig. 2). Individual cores are up to 1100 m long, but average about 300 m. Salts are well preserved and are sealed in double plastic covers which prevented decay and dissolution of the salts. These deposits were studied by core logging and X-ray diffraction was utilised to determine mineralogy. Bromine and potassium concentrations in the evaporites were obtained by the D.M.R. The bromine concentrations in the halite were determined by oxidation spectrophotometry using samples of about 2 g of halite and the analyses have 5% analytical error. A total of 100 thin sections were obtained from different evaporite lithologies for petrographic examination. Scanning

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

33

Fig. 1. Geological map of southeast Asia showing the general tectonic elements of the region and the location of the Khorat and the Sakon Nakhon basins.

electron microscopy and back-scatter imaging were used to define mineral textures and composition. The mineral analyses were performed at the centre of Microscopy and Microanalysis at the University of Western Australia in Perth, Australia. The δ34 S of 30 anhydrite samples were obtained from different cores and were analysed according to the methods of Holt and Engelkemeir (1970) at the CSIRO Laboratory in Sydney, Australia.

3. Geology The Khorat Plateau presently has a high escarpment of about 900 m above sea level along its western and southern edges whereas elevations of its

central area are only 100–300 m above sea level. Much of the sedimentary succession that makes up the Khorat Plateau is a part of an extensive, largely non-marine depositional system in the mainland of southeast Asia. It covers an area of 170,000 km2 of the Esarn region in northeastern Thailand and central Laos, and lies between latitudes 14º and 19ºN and between longitudes 101º and 106ºE (Fig. 1). The plateau is located on the Indochina microplate and includes the Sakon Nakhon Basin to the north and the Khorat Basin to the south. These two basins are separated by the Phu Phan anticline in northeastern Thailand. The plateau is a broad synclinorium bounded to the west by the Shan Thai microplate, and to the north by the South China plate. The Indochina microplate contains sediments ranging in

34

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Fig. 2. Relative location of some representative cores of the Maha Sarakham Formation in the Khorat and Sakon Nakhon basins.

age from Late Cambrian to Recent (Sattayarak et al., 1991; Mouret, 1994). Early to Middle Triassic collision between the Shan-Thai and the Indochina microplates along the Nan-Uttaradit suture (Sengor, 1979, 1984; Bunopas and Vella, 1992; Drumm et al., 1993), was followed by Late Triassic tectonic relaxation or extension which created half-graben basins, where thicknesses of up to 5 km of the non-marine red beds of the Khorat Group were deposited (Fig. 3). Fluvial and lacustrine facies filled these basins with conglomerates, sandstones, and mudstones that range in age from latest Triassic to late-Early Cretaceous (Racey et al., 1994; Mouret, 1994). The source of the Khorat rocks in Thailand is from erosion of the late Palaeozoic rocks exposed in the Nan-Uttaradit suture area in central north Thailand (Sengor, 1979; Hutchison, 1989) and possibly from eastern Laos and central Vietnam (Drumm et al., 1993). The Khorat deposits were intruded by granites of Campanian and Cenomanian

age which resulted from thermal subsidence of the Khorat Plateau (Smith et al., 1996). During the early Paleocene, compression from the northeast due to continental collision of the Indochina microplate with southeast China plate and from backarc compression of the Shan Thai microplate to the west, resulted in uplift and erosion of about 3000 m of the Khorat sediments and the formation of the NW–SE-trending Phu Phan anticlinorium in the central part of the Khorat Plateau (Cooper et al., 1989; Mouret, 1994; Bunopas and Vella, 1992). During the Cretaceous, the Indochina microplate was located near 20ºN latitude, suggesting arid climatic conditions (Achache et al., 1983). The Maha Sarakham Formation was first named by Gardner et al. (1967) and, based on palynomorphs, it is of Albian–Cenomanian age (Sattayarak et al., 1991). The formation averages 250 m thick and is up to 1.1 km thick in the centre of the Khorat Basin. The variation in thickness appears to be due

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

35

Fig. 3. Mesozoic and the Cenozoic stratigraphy of the Khorat Basin in NE Thailand and summary of the geological history of the Indochina microplate during the Mesozoic and the Cenozoic.

to: (1) differential subsidence before and during deposition; (2) localised tectonic differentiation, such as development of fault-controlled highs and lows; and (3) post-depositional dissolution, particularly of the Middle and Upper Units (due to very shallow burial). Lithological, stratigraphic and mineralogical similarities of the formation in both the Sakon Nakhon and Khorat basins, suggest that a single giant evaporite basin existed at least in the area of the present-day basins (Fig. 4).

4. Stratigraphy and depositional patterns The Maha Sarakham Formation comprises three distinctive depositional members (Lower, Middle and Upper) which are mainly composed of evaporites separated by red-coloured siliciclastics (Fig. 5).

All of the evaporative members contain beds composed of halite-replaced pseudomorphs of bottomgrowth gypsum. While such beds are common throughout both basins, these replaced beds are thickest and most plentiful at the southwestern corner of the basin. The following section of the paper is a petrographic description of all three members and their component units. Because they all share common diagenetic and geochemical histories, the depositional and diagenetic background for all of the facies will be reserved for the discussion section. The original depositional morphology of the Mara Sarakham has been complicated by deformation. Seismic data show that salt structures such as domes, anticlines, ridges and basins are found in the subsurface of the Khorat plateau (see Figs. 16 and 17). Circular mounds and shapes of rounded landforms are related to shallow salt domes and some of these

36

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Fig. 4. Idealised map showing the predominant rock types of the Khorat Plateau and the Maha Sarakham Formation. Large arrow indicates the postulated source of marine water inflow into the Khorat Plateau near Bamnet Narong area at the southwestern corner of the Khorat Plateau. KK marks the location of Khon Kaen City located in the central area of the Khorat Plateau.

salt structures are present as shallow as 50 m below the surface. These structures have apparently initiated from salt movement associated with differential loading of clastics after the deposition of the thick Lower Member. However, later stream channels have eroded up to 200 m of the post-salt sedimentary cover, permitting further movement along axes of salt domes and ridges. 4.1. Lower Member The Lower Member is well preserved throughout both basins. All of the component units are well displayed and are readily traced from core to core. Deformational thickening (and thinning) has affected this member more than the successive ones resulting in a great difference in thickness from area to area. The thickness of the formation ranges from 50 m in basin marginal areas up to 1100 m in the basin centre areas. 4.1.1. Basal Anhydrite Unit This unit is found at the base of the Maha Sarakham Formation throughout both the Khorat and

the Sakon Nakhon basins with a consistent basin-wide thickness of 1.1 m (Fig. 6). The unit does not appear to interfinger with or pass laterally into carbonates or clastics. The anhydrite exhibits laminar and nodular forms, and has a sharp and stylolitic basal contact with the underlying sandstones of the Khok Kruat Formation. Laminar and microcrystalline anhydrite is found just at the contact of salt and clastics. This type of anhydrite is made up of small and flattened anhydrite nodules of up to 2 cm size, forming even laminations. The upper contact of the anhydrite unit with the overlying halite is sharp and is marked by white anhydrite nodules found in thin layers. Generally, the anhydrite comprises poorly defined layers of hard, milky white to bluish-coloured nodules with a distorted and sheared mosaic fabric. 4.1.2. Halite L1 Unit The Halite L1 Unit is the most fully preserved and complete salt unit found in the Khorat Plateau. Thickness of this salt unit varies from 30 to 350 m. This unit is widely distributed in both basins and constitutes a single laterally continuous salt layer and

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Fig. 5. Lithostratigraphy of a complete sedimentary section of the Maha Sarakham Formation.

37

38

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Fig. 6. A composite core photograph showing the characteristic similarities seen in the Basal Anhydrite Member. The individual cores are obtained from different parts of the basin. This Basal Anhydrite Unit is traced for hundreds of kilometres and defines the base of the Maha Sarakham Formation. The unit averages 1 m in thickness. Length of individual core sample is 50 cm.

is present in all of the boreholes, but is somewhat deformed. Individual halite beds are relatively thin and average 15 cm thick. Lithologies of this unit include halite interbedded with minor anhydrite stringers and poorly preserved gypsum. In the lower section, the halite is almost pure (95%) and its crystals are nearly flattened, forming sheet-like halite crystals. The salt of this unit is dominated by sheared and recrystallised smoky or grey halite with numerous

well-formed, white, chevron halite beds (Fig. 7A). The chevron halite commonly contains fine clastic material ‘dusted’ along the chevron surfaces. The tops of chevron halite layers are marked by irregular dissolution surfaces, outlined by fine clastic grains, and are overlain and infilled by thin-bedded, clear halite making up bands of about 10 cm thickness. In some cases halite-replaced, bottom-grown, gypsum layers (up to 10 cm thick), overlie the clear halite

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

39

Fig. 7. Examples of halite from the Lower Member. (A) Core photograph showing sheared, massive and poorly bedded halite. An individual core segment is 1 m long. (B) Core photograph of halite with anhydrite nodules that define bedding. Core is 20 cm long.

layers. The depositional bedding of most of the halite is indistinct, particularly when halite contains little or no anhydrite. Where bedding is obvious, it is defined by isolated nodules or irregular masses of anhydrite stringers (Fig. 7B). 4.1.3. Anhydrite Marker Unit 1 This unit comprises the thickest anhydrite bed in all of the salt members and includes several layers

of laminated anhydrite and beds of well-developed gypsum pseudomorphs replaced by halite (Fig. 8). The unit is up to 3.45 m thick and separates the Halite L1 Unit, from the Halite L2 Unit and is defined by its unique thickness and wide distribution in the basin. Anhydrite layers are dense and commonly exhibit even lamination and bedding. Some layers have halite-replaced gypsum pseudomorphs in which outlines of the original gypsum crystals

40

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

are still in growth position (outlined by a rim of anhydrite) with a vertical orientation, as originally described by Hovorka (1992). Because of the anhydrite rims surrounding halite fillings, these layers appear to have an unusual sieve-like texture. 4.1.4. Halite L2 Unit The Halite L2 Unit varies from 10 to 200 m in thickness. This unit is characterised by thick (up to 1 m), white and clear salt beds with well defined bedding (Fig. 9). The crystals of halite in this unit are much less deformed than the lower section. The halite is dominated by chevron-type texture and is composed of white to creamy white halite that is rich in fluid inclusions that cause the milky colouration. However, halite crystals in these layers are smaller than those of the underlying Halite L1 Unit and in places exhibit a crystalline mosaic of anhedral texture. The tops of almost all of the chevron halite layers are truncated and marked by dissolution pits. The pits are filled by clear halite bands that average 3 cm in thickness. These clear halite layers are commonly overlain by the anhydrite stringers. Thin beds of halite-replaced gypsum pseudomorphs and laminated anhydrite are also present in this unit, making up to 8% of the rock volume and average 5 cm in thickness (Fig. 9A). Toward the top of the L2 Unit thin, millimetric, anhydrite beds are present, interbedded with the halite (Fig. 9B). 4.1.5. Potash Unit The distribution of potash deposits in the Khorat Basin is known from boreholes and seismics. Thickness and distribution of the Potash Unit is highly variable, possibly due to some depositional differences, but more importantly due to patchy dissolution and deformation. Core examination (of over 100 cores) shows that the potash deposits are extensive, seemingly laterally continuous, and in many areas are found at relatively shallow depths, at 100 m,

41

in both basins. The potash minerals include sylvite (KCl) and carnallite (MgCl2 ÐKClÐ6H2 O) which are found at the top of the Halite L2 Unit and are overlain by banded and red-coloured halite beds. On a macroscopic scale, the strata of the potash sequence include three zones: (1) a lower zone of massive halite with traces of carnallite filling dissolution cavities in halite; (2) a middle zone of massive and poorly bedded carnallite and halite; and (3) an upper sylvinite zone (interbedded sylvite, sylvinite, and halite) (Fig. 10). Carnallite is by far the most widespread potash mineral in the Khorat Plateau and its first appearance typically defines the base of the potash zone. It is highly variable in thickness, ranging from 10 to 80 m (average of 50 m). The lower part of the carnallite zone is composed of massive halite beds having irregular dissolution surfaces marked by traces of carnallite, collapsed and recrystallised halite and cavities coated with microcrystalline and red-coloured carnallite. This zone grades upward into massive and poorly bedded carnallite with a granular to subrounded texture. The middle part of the potash zone is dominated by massive carnallite and halite. This zone has an average thickness of 20 m and is poorly bedded. Two types of halite are observed in this zone: (1) large, clear, recrystallised, halite crystals up to 4 cm in size; and (2) granular and fine-grained halite. The first type is in the form of massive halite beds and the latter type is found as carnallitite including disseminated halite crystals in carnallite. Anhydrite and other less soluble residues are scarce in the carnallite zone. The carnallite is poorly bedded and in most cases is pale red or clear in colour and its crystals are granular, of up to 1 cm in size. In the uppermost part of the potash zone there is a sylvinite section (a mixture of halite and sylvite) that directly overlies the carnallite-rich zone. The average thickness of this zone is 4 m. The sylvinite

Fig. 8. Examples of Anhydrite Marker Unit (indicated by arrows). (A) Anhydrite layer is thick (about 80 cm) and includes ‘sieve-like’ textures that exhibit pseudomorphs of halite after gypsum and thin laminar anhydrite layers. Individual core segment is about 0.8 m long. (B) A complete section of a marker bed of about 0.5 m thick, interbedded with halite and other minor anhydrite beds that define bedding with halite. Individual core segment is about 100 cm long. (C) An example of a thick anhydrite marker bed which is 2.2 m thick and is composed of alternating laminar anhydrite and gypsum pseudomorphs. Individual core segment is 1 m long. The well defined gypsum forms suggest that the original gypsum crystals that nucleated at the bottom of the salt pan were replaced early on in the diagenetic history of the deposits as synsedimentary diagenetic process.

42

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Fig. 9. Typical sedimentary characteristics of bedded halite as seen in the upper section of the Maha Sarakham Formation. (A and B) Bedding of halite is defined by thin and irregular anhydrite stringers. Dark bands in halite beds include fine clastics within clear halite. The dark bands form during flushing of the halite salt pan by fresh water. Core segment is 1 m long.

zone begins at the top of the carnallite=halite which grades upward into sylvite or sylvite=halite. Each sylvite-rich bed ranges from 1 to 15 cm in thickness. Macroscopically, sylvite layers are not homogeneous and several forms of sylvite are commonly found both intermixed and as thin but separate layers with fine clastics and thin halite beds (Fig. 11). These thin halite layers still exhibit well-defined bedding

and in many places display chevron halite textures (i.e. are primary). Some of the halite associated with sylvite, particularly in regions of structural deformation, is coarsely crystalline and has a marked blue colour. Blue-coloured halite is poorly crystalline and is found in zones of up to 10 m thick. Sylvite is present as: (a) individual crystals of up to 1 cm size of euhedral forms, found in carnallite or halite

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62 Fig. 10. Sylvite lithologies. (A) A core photograph of massive, red-coloured, bedded sylvite and halite. Core sample is 12 cm long. (B) Core photograph showing bands of massive halite and sylvite patches or clusters. Sylvite crystals show a characteristic ‘amoeboid’ texture. Individual sylvite crystal is up to 1 cm in size. Core sample is 12 cm long. (C) A core photograph of massive sylvite. Sylvite crystals are showing interlocking fabrics. Sylvite crystals are coated with thin clay seams. Core is 15 cm long. 43

44 M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62 Fig. 11. Microfabrics of the potash zone. (A) Large and euhedral sylvite crystal at the centre surrounded by carnallite (C). Scale bar is 1 mm. (B) Thin layer of sylvite (S) between halite (H). Scale bar is 1 mm. (C) Large and euhedral sylvite crystals (S) surround subhedral halite (H). Scale bar is 1 mm. (D) Anhedral sylvite crystals (S) cementing euhedral halite crystals. Scale bar is 1 mm.

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

(Fig. 11A); (b) massive lenses or layers that average 3 cm in thickness or interlayered with halite (Fig. 11B); (c) small clusters and groups of crystals within carnallite or halite (Fig. 11C); (d) irregular patches within massive halite beds (Fig. 11D); and (e) massive beds with an average thickness of 10 cm. Petrographically, sylvite crystals exhibit well crystallised subrounded shapes and contain finely disseminated haematite inclusions. Traces of anhydrite and thin beds of gypsum pseudomorphs composed of halite are present throughout the sylvite-rich zone. Associated with sylvite and halite are accessory borate minerals of hilgardite [Ca2 BCl (OH)2 ] and boracite (Mg3 ClB7 O13 ). These minerals occur in two forms: as white thin layers up to 1 cm thick in sylvinite beds or as dispersed massive and irregular nodules and grains from 1 mm up to 5 cm thick. Euhedral tachyhydrite (CaCl2 .MgCl2 .12H2 O) crystals are commonly found in these potash-rich deposits. The identification of tachyhydrite was confirmed by X-ray microanalysis with scanning electron microscope (SEM). Tachyhydrite crystals are small and average 1 mm in size and are commonly found with halite and carnallite (Fig. 12), but is not detected with sylvite. 4.1.6. Clastics L Unit The thickness of this unit is 10–60 m. The Clastics L1 Unit is a mudstone of reddish to brown colour that directly overlies the Lower Member (above the halite and potash-rich units). The mudstone layers are mainly composed of illite, quartz, haematite with minor K-feldspar. These layers are massive and are interbedded with rare siltstone layers. Displacive halite hoppers of up to 3 cm size occur in these red mudstones, and chaotic mudstone textures composed of clastic matrix with irregular masses of halite skeletal crystals are common. Randomly oriented fractures filled with halite spar of up to 1 m thick are found in this unit. A few root structures and weakly developed soil profiles have been noted; however, the disrupted and chaotic textures are the dominant fabric. 4.2. Middle Member The Middle Member ranges from 40 to 130 metres in thickness, is composed of two halite units

45

separated by an anhydrite marker bed and contains most of the same depositional and diagenetic features as the Lower Member. There is little or no potassic salt in this member. 4.2.1. Halite M1 Unit The thickness of this halite unit is 20–60 m. This unit directly overlies the Clastics L Unit, and is composed of well-bedded halite layers that average 10 cm in thickness. The halite beds are interbedded with thin beds of anhydrite and layers of gypsum pseudomorphs that are composed of halite with typical anhydrite rims. Some disseminated sylvite and carnallite crystals are found in rare intervals of this sequence. Halite beds in this unit typically exhibit coarse chevron structures and have a characteristic dark honey-colour, due to finely disseminated iron oxide. 4.2.2. Anhydrite Marker M This marker bed is extensive and contains abundant well-formed gypsum pseudomorphs. The unit is up to 0.5 m thick, and is similar to the unit found in the Lower Member. It consists of beds composed of one or more thin, laminated anhydrite layers of up to 10 cm thick immediately overlain by more massive beds composed of gypsum pseudomorphs. The gypsum is replaced by halite and the original gypsum crystals still show original vertical orientation as defined by anhydrite rims. 4.2.3. Halite M2 Unit The thickness of this halite unit is 10–70 m thick and is separated from the Lower Halite M1 Unit by the Anhydrite Marker Bed. This unit is largely composed of bedded dark honey-coloured rock salt, interbedded with dark smoky-coloured halite beds. Thin anhydrite layers of up to 5 cm thick are present. These layers are composed of anhydrite nodules, laminated anhydrite, and halite-replaced, gypsum pseudomorphs with anhydrite rims. 4.2.4. Clastics M Unit The thickness of this unit is 20–70 m. Clastics consist of massive red to purple claystone and silty mudstone. The bedding in this unit is well defined, with laminations and root traces and beds up to 50 cm thick. Internally the clastic beds are highly

46

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Fig. 12. Microfabrics of halite and tachyhydrite. (A) Thin-section photomicrograph of euhedral tachyhydrite crystals (T) found in blocky halite. Note that the tachyhydrite crystals are well formed. (B) SEM photomicrograph of euhedral tachyhydrite (T) in halite (H). Scale bar is 0.1 mm.

fractured and are poorly consolidated. At the base of the unit, near the contact of halite and clastics, characteristic grey anhydrite layers of up to 1 m thick

are present. Displacive cubes of halite crystals and nodules of anhydrite of up to 4 cm size are found dispersed in the fine clastic matrix.

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

4.3. Upper Member The entire sedimentary section of the Maha Sarakham Formation was sampled by 235 cores but only about 12 cores contained the Upper Member, as it is poorly preserved in the rock record largely due to its shallow burial in a wet climate. Based on what halite is now preserved in this unit, it seems that at least half of the salt of this unit has been leached. Where preserved the total thickness of this member ranges from 3 to 17 m. 4.3.1. Halite U1 Unit The thickness of this unit ranges from 1 to 15 m. This salt unit is the least preserved halite unit in the Khorat Plateau because of later dissolution and possible non-deposition in some parts of the basins. However, salt beds when observed are the least deformed in the sequence. This unit is mainly composed of layers of chevron halite that average 5 cm in thickness. Halite is dark smoky grey to orange in colour with minor anhydrite stringers and some layers of gypsum pseudomorphs. 4.3.2. Anhydrite Marker Unit U The thickness of this anhydrite unit where present is up to 0.5 m. This unit is composed of thin beds of laminated anhydrite and layers of gypsum pseudomorphs that are composed of halite with an anhydrite outline. 4.3.3. Halite U2 Unit The thickness of this unit is 2 m. This unit is rarely sampled by the studied cores due to later dissolution. Where present it contains minor anhydrite stringers with thin carbonaceous bands that are interbedded with the halite layers. Halite of this unit is well bedded and is dark smoky to orange in colour with beds up to 10 cm thick. 4.3.4. Clastics U Unit This unit represents the uppermost sedimentary layer of the Maha Sarakham Formation. The thickness of this unit is highly variable, up to 680 m. The clastics of this unit are composed of pale reddish-brown silty claystones and sandstones. Bedding is well defined with even laminations, and sets of cross-beds are commonly observed. Some layers contain well-

47

defined root traces. Sands are mainly composed of detrital quartz grains of up to 2 mm in diameter but minor amounts of feldspar are present. The detrital grains are rounded to subrounded. Composite sets of cross-beds of silty and sandy mudstones interbedded with massive mudstone layers occur in this unit. The contact of this unit with the clastics of the overlying Phu Thok Formation is not clearly defined; however, generally the clastics of the Phu Thok are coarser. 4.4. Sedimentology synthesis 4.4.1. Primary depositional facies Sulphates. Evaporated marine water, upon reaching a salinity range of 140–300 ppt, results in the formation of crystalline gypsum at the bottom of the salt water body. Schreiber and Kinsman (1975) and GeislerCussey (1982) describe the gypsum morphologies of deposits forming in shallow environments and note a number of different gypsum crystal forms, but the most common primary crystal forms are twinned gypsum crystals, commonly present in regular, coalescent beds. Such beds make up much of the Upper Miocene gypsum present all around the Mediterranean. Commonly when such gypsum deposits become buried to depths greater than 1 km they gradually lose their water of crystallisation and become massive, featureless beds of anhydrite (Jowett et al., 1993). The characteristic gypsum crystal morphology is usually destroyed with the 40% volume loss and is preserved only under unusual circumstances (Schreiber et al., 1982). Pseudomorphs of such shallow-water gypsum beds (see ‘diagenesis’ below) are found throughout the Maha Sarakham Formation and are particularly common in almost every bed near the southwestern corner of the Khorat Basin. Several other morphologies of calcium sulphate evaporites are found in the Maha Sarakham Formation (Fig. 13). Beds of laminar and microcrystalline anhydrite are commonly found within the M and U clastic units. Isolated nodules of anhydrite are also found within these clastic units. Well bedded and crystalline anhydrite that exhibits a nodular-mosaic (‘chickenwire’) texture is found in clastics at the western area of the Khorat Basin, in the Bamnet Narong area. These beds are found in sections of up to 15 m thick. This type of anhydrite is white and is found in thick beds of nodular anhydrite that a exhibit mosaic texture. Shear-

48

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Fig. 13. Examples of different calcium sulphate lithologies in the Maha Sarakham Formation. (A) Pseudomorphs of gypsum crystals now are replaced by anhydrite. The original ingrowth morphologies suggest that these forms are still in their crystallisation position. Sample is 15 cm long. (B) Beds of anhydrite that exhibit twin gypsum forms. Sample is 20 cm long. (C) Small and poorly bedded anhydrite nodules in composite mosaic forms. Core is 20 cm long. (D) A core photograph showing numerous anhydrite nodules.

man (1978) and Butler et al. (1983) have demonstrated that such nodules and layers form displacively within an already existing subaerial matrix adjacent to a marine water source, as early diagenetic features. Veigas (1997) and Ortı´ (1997) show that many of the same very early displacive morphologies may form on the floors of and adjacent to sporadically desiccated playa lakes. Because of the morphological similarities between nodular calcium sulphates, in many cases, the only way to differentiate between marine and continental-sourced sulphates is by examination of the sulphur isotope values. Halite. Halite beds having chevron structures were first described by Shearman (1970). Shearman

(1970) pointed out that these structures develop during rapid growth of halite crystals at the bottom of shallow salt pans, and that their milky colour is caused by fluid inclusions incorporated on the growth faces of the salt during rapid crystal growth (Shearman, 1978). Truncation and erosion of upper surfaces of these beds (including solution pitting) results from short-term subaerial exposure, and=or influx of new, less concentrated water (below saturation). Infill of such solution pits and surface irregularities by clear, banded halite suggests flooding by new water and renewed concentration of this ensuing batch of water covering the surface. Lowenstein and Hardie (1985) show the types of halite formed in deeper water and contrasts them to the shallow-water

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

forms. The Maha Sarakham Formation contains repeated beds of shallow chevron halite beds and few examples of deeper-water halite; therefore its origin is largely from very shallow waters. 4.4.2. Diagenetic overprints Hovorka (1992), noted that early replacement of gypsum crystal beds by halite (and anhydrite) may take place and the resultant pseudomorphs clearly retain the characteristic forms of the original shallowwater gypsum. Replacement of the bottom nucleated gypsum actually occurs in the floors of salt pans when hot brines (33–37ºC or greater), that are undersaturated in gypsum and oversaturated in halite, sink into the previously accumulated gypsum substrate (Schreiber and Walker, 1992). Schreiber and Walker point out that it is the hot, halite-saturated brines that heat the now cooler underlying gypsum, dissolve it, and as the sinking brines begin to cool, replace most of the gypsum with halite. The anhydrite outlines of the original gypsum crystals remain preserved, floating in the replacive halite (Hovorka, 1992). The Basal Anhydrite Member found at the base of the Maha Sarakham Formation is defined as a dissolution residue that resulted from flushing of the salt by compacted basin waters. This process resulted in accumulation of interbedded anhydrite with the salt as a dissolution residue at the base of the salt. Lithologic, petrographic and geochemical data supporting this interpretation are given in El Tabakh et al. (1998). Tectonic effects. In many halite beds of the Lower and Middle Members, the halite crystals appear to be strongly deformed, even where the beds themselves still reflect primary depositional features. Sheared and flattened crystals are common. Salt deformation on structure has been recognised for a long time (Clabaugh, 1962), but experimental studies of halite recrystallisation, resulting in foliated fabrics, did not take place until much later. Investigation of salt domes as hydrocarbon storage facilities resulted in studies of experimentally induced crystal changes developed under fairly low strain rates (Larsen, 1985). As a consequence of these investigations it is clear that pervasive recrystallisation can take place even where the beds are only modestly

49

deformed. The original geometry of particularly the sylvite deposits is not defined in the Maha Sarakham Formation. This is because later extensive dissolution and alteration by groundwater and deformation of salt occurring during development of salt domes, ridges and depressions, have drastically changed the geometry of the potash deposits and also the rock salt. The surface extension and vertical changes of thickness of the sylvite and carnallite associations greatly change from one core to another. Extension and vertical changes of thickness of the potash greatly change from one core to another, even on the scale of hundreds of metres between cores. Other features. Blue-coloured halite is commonly found overlying deformed zones of sylvite (and sylvinite) in the Maha Sarakham. Sonnenfeld (1995) has pointed out that zones of blue-coloured halite commonly develop along paths of circulating brines in association with deposits of sylvinites. He further suggests that the colour is not due to impurities within the salt but is probably due to recrystallisation of the associated halite and the loss of bromine atoms from the crystal lattices, leaving metallic sodium behind. This type of halite has probably resulted from fluid migration resulting from stress during doming of the salts and is not a primary feature. The effects of dissolution of the Maha Sarakham salts are obvious in the observed thinning of salt units and the absence of units in several cases, particularly around the basin edge. Dissolution of salt leads to lack of preservation of the Middle and Upper salts in many cores. It also resulted in accumulation of anhydrite residue from dissolution of salt in some beds. This residue is found between the overlying clastics and the underlying salt unit (Fig. 14A). Typical anhydrite-dominated thin residual layers tend to cap underlying salt beds. These anhydrite residues tend to follow modern hydrological and topographic patterns. The contacts between the residue and the original halite beds show characteristic angular discontinuities (Fig. 14B).

5. Geochemistry of the evaporites The concentration of bromine in halite and isotopic composition of δ34 S in anhydrite are conven-

50

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Fig. 14. Dissolution features of Maha Sarakham salt. (A) Core photograph showing horizontal, stacked, light anhydrite bands overlying inclined halite beds with thin layers of anhydrite stringers. (B) A core photograph of anhydrite layers of about 2.5 m thick found between halite and clastics. The anhydrite bands apparently are a residue from dissolution of salt layers along tops and flanks of salt domes.

tionally used to help define the origin of the brines from which the evaporative sediments have formed. These studies cannot stand alone but must be compared to the petrology of the sediments from which they are taken.

5.1. The Br composition in the salts The bromine content of halite may be used as a general test for the origin of the water from which the halite was formed. This is a useful measurement because in nature bromine may substitute, in part, for chlorine in the halite crystal lattice. The amount

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

of bromine substitution for chlorine depends on the concentration of bromine in the original brine. Because the bromide ions have some difficulty in fitting into the chloride positions in the crystal lattice of the halite, the percentage of bromine entering into the crystal growth is lower than its percentage in the solution. Thus more and more bromine remains in the parent solution as salt is precipitated. The concentration of bromine in halite is therefore used to determine brine evolution during halite precipitation (Holser, 1979). The first halite precipitates from evaporation of seawater will contain 65–75 ppm Br and will progressively increase during further halite deposition rising to values of 320 to 400 ppm at the point where potash minerals should precipitate from evaporated seawater (Holser, 1966, 1979). In the Lower Member, the initial bromine content at the base of halite averages 70 ppm (Fig. 15). The bromine content of this unit shows a slow but continuous increase from the bottom (70–90 ppm), to the middle (200 to 250 ppm), to the top where it rapidly increases to 450 ppm (just below the potash layer). In some cores, Br concentration in the Lower Halite Member is constant from the base to the top and averages 75 ppm. In the Middle Member the initial concentration of bromine is 200 ppm at the bottom, and increases rapidly to 400 ppm in the middle, reaching up to 450 ppm in the upper part of this member. The bromine concentration in the Upper Member exhibits uniform an upward increase and ranges from 180 up to 400 ppm. However, bromine concentration may retreat to about 80 ppm in the middle of the Upper Salt and then increase rapidly. 5.2. The Ž34 S isotope values in anhydrite The evolution and variation of the sulphur isotopes in marine evaporites has been documented by a world-wide analysis of evaporites of different ages (Claypool et al., 1980). In order to determine the δ34 S isotopic values of anhydrite in the Maha Sarakham Formation 30 anhydrite samples were taken and analysed from the different types of anhydrite (Table 1). These values are used to trace the origin of aqueous sulphate from which evaporites precipitated. The δ34 S isotopic composition in anhydrite taken from the halite beds ranges from 14.8‰ to 17.7‰ (CDT). The δ34 S isotopic composition in an-

51

hydrite nodules taken from the thick clastic units shows significantly lower isotopic values, ranging from 6.4‰ to 10.9‰ (CDT). 5.3. Discussion of geochemical studies Geochemical analysis of bromine in the halite of the Maha Sarakham Formation demonstrates the following: (1) concentration of bromine in the Lower Member ranges from 70 ppm to 400 ppm (base to top), that more or less conforms with the evaporative evolution of marine water as pointed out by Hite and Japakasetr (1979); and (2) bromine systematically increases in each of the three members and can be traced from one core to another. Bromine concentration and upward systematic increase in all halite members across both basins demonstrates excellent lateral continuity, and that both basins were connected during deposition. Bromine data for both the Middle and Upper Halite Members suggest recycling and dissolution=reprecipitation of the salt. This interpretation is also supported by Hite and Japakasetr (1979). Data presented here and those of Hite and Japakasetr (1979) show Br curves for the Middle and Upper Halite Members that suggest synsedimentary recycling and dissolution=reprecipitation of the salt. The concentration of bromine in the Maha Sarakham Formation indicates the following: (1) Br is initially formed from marine water as in the base of the Lower Salt Member; (2) Br is recycled as in Middle and Upper Salt Members; and (3) there are internal variations in the concentration of Br in the Upper Salt Member. Later chemical reworking or dissolution of halite may deplete the original salt in bromine as invoked by Hite and Japakasetr (1979) for the pattern of Br curves in the Maha Sarakham Formation. However, early chemical fractionation in a stratified water body is preferred because it may cause enrichment of the remaining brine in bromine. This same mechanism is suggested by Cendon et al., 1998 for a synsedimentary origin of sylvite deposits of Subiza in Navarra, Spain. The Mesozoic era was a time of world-wide formation of evaporative basins leading to significant isotopic variations in marine evaporites, but the sulphur isotope values of anhydrite samples taken from the Maha Sarakham salts are in agreement

52

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Fig. 15. Br profiles of different salt units of the Maha Sarakham Formation. The bromine concentration increases upward in most of the halite units suggesting its early sedimentary recycling.

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

53

Table 1 The isotopic composition of sulphur in different anhydrite forms in the Maha Sarakham Formation Unit

Core

δ34 S

Depth (m)

Description

Upper Clastics Upper Clastics Upper Halite Middle Halite Middle Halite Lower Clastics Lower Clastics Top L. Halite Top L. Halite Top L. Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Lower Halite Basal Anhydrite Basal Anhydrite Basal Anhydrite Basal Anhydrite

k-95 k-36 k-49 RS 2.7 k-11 k-62 k-62 k-31 k-91 RS 2.22 k-14 kk-03 k-49 k-3 k-16 RS 2.7 RS 2.1 k-47 k-64 k-11 k-40 k-104 k-56 k-8 k-19 RS 2.12 k-29 k-16 k-19 k-56

8.6 10.9 15.4 15.9 16.2 6.4 8.6 14.3 15.1 15.9 15.4 15.5 16.0 16.0 15.8 17.0 16.9 16.0 16.2 16.1 15.9 14.7 16.9 14.8 15.1 17.7 16.0 15.1 15.4 15.0

264 184 63 254 154 410 212 318 108 93 78 265 81 63 34 211 192 96 266 490 500 240 204 157 224 139 245 359 224 365

Anhydrite nodule in clastics Anhydrite nodule in clastics Anhydrite residue in halite Anhydrite nodule in halite Anhydrite nodule in halite Anhydrite nodule in clastics Anhydrite nodule in clastics Anhydrite cap=residue Anhydrite residue=cap Anhydrite residue=cap Anhydrite nodule in halite Laminar anhydrite in halite Laminar anhydrite in halite Anhydrite nodule in halite Anhydrite marker in halite Anhydrite marker in halite Anhydrite residue in halite Anhydrite marker in halite Anhydride in halite Anhydrite residue in halite Anhydrite nodule in halite Anhydrite nodule in halite Anhydrite in a marker bed Anhydrite nodule in halite Anhydrite nodule in halite Anhydrite bed Anhydrite bed Anhydrite bed Anhydrite bed Anhydrite bed

Data show marine values of anhydrite taken from halite and from anhydrite residues in halite. However, anhydrite nodules in red-coloured clastics are of low sulphur isotopic values, suggesting non-marine origin of these evaporites.

with world-wide Cretaceous marine evaporite values which range from 14‰ to 17‰ (CDT). The thick clastic units, that ended each of the three marine evaporite phases, exhibit sedimentary features characteristic of non-marine fluvial to lacustrine environments. The sulphates present in these clearly nonmarine units do not fit the word-wide marine curves (6.4‰ to 10.9‰ (CDT)) and are appropriately the product of continental or mixed-water precipitation.

6. Sedimentology and stratigraphic correlation Several factors have contributed to the formation and preservation of the salt units of the Maha

Sarakham Formation such as geographic isolation, arid climate, and a strong supply of solutes. Lithologies typical of evaporative basin margins such as evaporitic carbonates, dolomites, and reefs are not known from the Khorat Plateau. This is possibly due to (1) limited sampling, (2) non-deposition, or (3) subsequent erosion of these lithologies. The primary fabrics in halite beds are composed of inclusion-rich chevron halite, halite-replaced gypsum (vertically oriented) and equant halite (recrystallised). These textures suggest deposition in a shallow salinepan environment (Shearman, 1970; Lowenstein and Hardie, 1985; Hovorka, 1992). These beds are crosscut by abundant dissolution surfaces filled with a coarsely crystalline cement of clear halite. Bottom-

54

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

grown gypsum crusts and anhydrite together with numerous dissolution surfaces suggest influx of dilute waters of possibly marine origin in salinas of shallow brine depths (Warren, 1982). The dilution was, however, largely marine, based on isotopic values of the sulphates; however, continental and even hydrothermal waters may have entered the basin. Stratigraphic correlation of different units of the Maha Sarakham Formation is based on lithological and depositional features seen in cores and there is little evidence of fauna and=or flora. A typical cross-section of the Maha Sarakham Formation in the Khorat Basin is shown in Fig. 16. The thickness of the formation increases toward the central basin area. In all cases, halite and sandstone units increase in thickness and reach their maximum thickness in the central basin area. Siliciclastics in the basin area are composed of alternating cross-bedded siltstones and massive mudstones of fluvial origin suggesting fluvial deposition in the basin province. Lithologies and stratigraphy of the Maha Sarakham are relatively uniform over vast areas of both basins, suggesting stability of deposition as seen in seismic lines (Fig. 17). The continuity of the evaporites was disrupted by later salt doming and dissolution, causing relative thinning of different units at different parts of the basin. The stratigraphy and continuity of the salt strata which typify many of the saline giants of the world was confirmed by coring along and near the seismic lines. Gypsum layers and anhydrite partings lack reworking and suggest that no major phases of halite dissolution took place during freshening events (no notable synsedimentary collapse structures). This is also supported by a lack of significant insoluble residue between halite layers. Only thin dustings of clastic grains are present anywhere in the many halite beds. Depositional conditions suitable for sulphate and halite formation and preservation were disrupted by three periods, marked by deposition of siliciclastics and subaerial exposure evidenced in the three major mudstone interbeds within halite and associated halite=mudstone mixtures. Sedimentary structures in mudstones such as mudcracks, chaotic textures, root structures and burrows suggest a subaerial setting. However, haloturbation caused by early diagenetic displacive growth of halite in mudstone beds suggests

sporadic brine flooding and then drying of subaerial exposed areas. Input of clastics into the basins implies that the source areas were exposed to similarly sporadic variations in weathering, erosion and rainfall.

7. Discussion 7.1. Depositional model Several depositional processes are interpreted from the study of the Maha Sarakham Formation: (1) marine-derived flooding onto the Khorat Plateau; (2) evaporation of seawater associated with aridity and hydrological restriction; (3) synsedimentary dissolution of salt beds; (4) basin-wide and repeated influx of marine water (freshening) that is evidenced in the deposition of thick sulphate marker beds which are now seen as largely replaced by halite. The thickness of this evaporite accumulation, particularly the potassic salts suggests extreme aridity that favoured precipitation and preservation of halite and potash minerals (Kinsman, 1976). Despite the fact that this seems to be a marinesourced deposit based on bromine data and sulphate isotope studies, the evaporites studied here do not have a regional setting along a typical passive continental depositional border. Based on available lithological data, we suggest that the location of the marine entrance into the Khorat Basin was in the Bamnet Narong area which is located in the southwestern part of the Khorat Plateau and deposition of the Maha Sarakham saline giant took place in a large inland and isolated basin which was separated from the Cretaceous ocean by a barrier located in the southwestern corner of the Khorat Plateau. This is because large amounts of calcium sulphate evaporites comprising anhydrite and secondary gypsum are found interbedded with red siliciclastics in the southeastern corner of the plateau. The deposition of these salts coincides with a world-wide high sea level in the Late Cretaceous and spill of marine-sourced water into what is now the Khorat Plateau (Haq et al., 1987). Other examples of saline giants in the world suggest several models of deposition of evaporites which include: (1) shallowing of water in a deep basin, as in the Mediterranean (Hsu¨, 1972); (2) drops in sea level be-

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62 Fig. 16. A west–east cross-section taken across the Bamnet Narong area at the southwestern corner of the Khorat Basin showing progressive increase in thickness of salt toward the centre of the basin. Anhydrite and gypsum are abundant in several cores at the western part of the section, near the basin’s western margin. 55

56 M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62 Fig. 17. Subsurface structures of the salt domes in the Maha Sarakham Formation. (a) A shallow seismic line showing a salt dome. The salt members are relatively thinner on top of the dome and are progressively thicker away from the dome due to dissolution by groundwater. Line is obtained from Khon Kaen area and the relative location of the line is given in Fig. 2. (b) A deep seismic line showing several salt domes at the subsurface of the Khorat Plateau. Notice the continuous pattern of deposition of the salt members as well as the strata of the Khorat non-marine clastics. Line is obtained from the near central area of the Khorat Basin and the relative location of the line is given in Fig. 2.

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

low the marginal topographic highs as in the Paradox Basin (Williams-Stroud, 1994); (3) seawater spilling into opening rifts from surrounding oceans as in the early Mesozoic basins of the Atlantic passive margins (Burke, 1975); (4) barred marginal marine to ephemeral salt pan (Williams, 1991); and (5) restriction and drawdown of a marine-sourced basin located entirely on continental crust as in the Michigan Basin (Cercone, 1988). Deposition of evaporites of the Khorat Plateau seem to have been controlled by repeated drops of sea level below the marginal topographic highs located in the southwest corner of the plateau in a depositional area that was located entirely on continental crust. 7.2. Potash mineralisation The origin of potash minerals is controversial because no modern potash-forming basins of marine origin are active in recent time. Modern potash deposits typically form from non-marine-sourced waters such as Lake Qaidam Basin, China (Lowenstein et al., 1989; Casas et al., 1991), Chatt El Djerid, southern Tunisia (Bryant et al., 1994), and Danakil Depression, Ethiopia (Holwerda and Hutchinson, 1968). The chemistry and mineralogy of halite and anhydrite of the Maha Sarakham Formation suggest a marine origin of these evaporites. However, the mineralogy of the potash sequence suggests a possible non-marine influence on mineralisation or at least recycled salts that resulted from dissolution of originally marine salts by non-marine water. A primary origin of sylvite formed by accumulation of crystals precipitated at the brine–air interface was suggested by Lowenstein and Spencer (1990) for the sylvite deposits of the Rhine graben, Permian Salado and Devonian Prairie formations. Textures of the potash minerals in the Maha Sarakham Formation suggest primary and=or early diagenetic origin including: (1) presence of carnallite at the base of the potash zone as fillings of dissolution cavities in halite beds; (2) lateral continuity and well-defined layering of sylvite and presence of interbedded primary chevron halite with the sylvite beds; and (3) fillings of sylvite in early dissolution cavities of halite layers. During deposition, once carnallite was formed as a near-surface deposit, it may dissolve when fresh

57

or nearly concentrated brine enters the basin. Subsequently sylvite will form as a primary deposit during these reactions. In the Khorat Plateau a substantial amount of carnallite is still preserved and sylvite exhibits primary textures suggesting that if carnallite to sylvite conversion occurred, it was minor. Evidence of freshening in the Khorat Plateau during sylvite formation includes: (1) presence of thin clastic beds composed of fine detritus and clays associated with sylvite; (2) presence of traces of anhydrite and mmsize gypsum pseudomorphs, replaced by halite; (3) presence of both chevron and clear halite beds interbedded with sylvite beds; and (4) presence of iron oxides in both carnallite and sylvite which were leached from red-coloured clastics, derived by incoming non-marine waters into the basin. New brine would fill pore spaces in carnallite and halite beds and original pore-filling brine would seep away. In the modern Qaidam Basin in China, carnallite and halite sediments exhibit porosity and permeability which allowed less saline and fresh water to enter the carnallite sediments (Casas et al., 1991). Due to dilution, carnallite would decompose into sylvite and MgCl2 dissolved in solution (Braitsch, 1971; Richter-Bernburg, 1972; Ortiz and Mur, 1984). Water released from the structure of carnallite would then seep down and react further with carnallite and convert it into sylvite. Thickness of sylvite varies from one core to the another that mainly depends on the original permeability of carnallite and chemistry of original potash-forming brine and the new brines that resulted from a carnallite–sylvite transformation. Brines rich in magnesium chloride end-member would have been produced in the potash deposits as residual brine and would escape by seepage through the deposit or become diluted by overflow brine (Spencer, 1983; Sonnenfeld, 1984). The potash deposits of the Maha Sarakham Formation lack MgSO4 -bearing salts. Final evaporation of modern seawater fails to form MgSO4 -bearing salts that should form before K and Mg-chloridebearing salts, e.g. polyhalite, kainite, kieserite. Several reasons can be given for the absence of these salts: (a) unusual marine waters, (b) modification of original mineralogy, (c) mixing of marine C nonmarine parent waters, (d) non-marine origin, and (e) partial removal of sulphate by sulphate-reducing bacteria (Braitsch, 1971). The Stassfurt evaporite

58

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

minerals of the Permian Zechstein are the only evaporite suits which appear to follow the predicted evaporation of modern seawater at 25ºC (Harvie et al., 1982; Lowenstein and Spencer, 1990). The formation of MgSO4 -deficient potash evaporites is possible in non-marine basins which contain CaCl2 -rich brines such as in the Chatt el Djerid depression, southern Tunisia, Qaidam Basin in China, and the Danakil Depression, Ethiopia. In these areas, marine-like halite, gypsum and anhydrite are commonly found associated with potash deposits. Magnesium sulphate-poor evaporites occur in cratonic sag basins where rising deep-seated hydrothermal brines are not expected. However, the hydraulic head, produced during the desiccation of deep evaporite basins, may cause deep and hot groundwaters to migrate upward and enter the basin (Kendall, 1989). Usually in an evaporative deposit originating from marine water, there is little calcium remaining in water concentrated above 325 ppt (start of halite precipitation). In this basin, because of early halite replacement of primary gypsum, CaCl2 -rich brines enter the waters of the basin and modify the chemistry of the surface waters and of the associated groundwater. Minerals associated with potash rocks can then be used to infer the paragenesis of the potash sequence. In the Maha Sarakham Formation tachyhydrite is abundant in the potash zone and although it is very soluble it is present in considerable quantity in the deposits. Here, tachyhydrite is commonly found associated with carnallite as alternating bands or as a mixture with carnallite. Core examination indicates that tachyhydrite is more concentrated in the central part of both basins. These areas were the most subsident parts of the basins, and subsidence allowed brines to flow into low areas. Tachyhydrite dissolves in contact with air because it is highly deliquescent and its presence in the Maha Sarakham evaporites suggests that these evaporites were deposited subaqueously or as a subsurface product of diagenesis by migrating groundwater. Based on geochemical data obtained from evaporation of seawater of different pathways and compositions, the mineral tachyhydrite is conclusively proven not to form from the final evaporation of pure marine-sourced waters (Hardie, 1990). CaCl2 -rich waters were described in association with potash deposits and may form in geothermal areas (Holwerda

and Hutchinson, 1968). The study by Hardie (1990) suggested that hydrothermal input into the potash precipitating environments must have occurred to account for the formation of tachyhydrite. In the Khorat Plateau, there is some evidence of hydrothermal activities and thermal event during the Cretaceous which have possibly supplied CaCl2 -rich waters into the basin (Smith et al., 1996). Dating by K–Ar of granite intrusions in the area gave ages of Cenomanian and Aptian that are coincident with the thermal subsidence of the Khorat Plateau. The timing and mechanism of the input of these hydrothermal sources is still not fully understood. However, these observations suggest an origin and=or contribution of waters from non-marine or thermal sources into the depositional area. Borate minerals are common in the potash-rich beds of the Maha Sarakham Formation and may result from non-marine input into the basin such as in the Michigan Basin (Nurmi and Friedman, 1977), the middle Oligocene lacustrine sediments which are interlayered with volcanic sediments and flows (Kyle, 1991), and in the Sierra Nevada lakes of the long Valley Caldera (Felmy and Wear, 1986). 7.3. Comparison to other potash basins The depositional characteristics of the Maha Sarakham saline deposits are comparable to other Cretaceous saline deposits, including deposits of the Early Cretaceous Congo–Gabon basins (deRuiter, 1979), and deposits of the Sergipe in Brazil (Wardlaw and Nicholls, 1972). These characteristics are: (1) including large volumes of tachyhydrite; (2) stratigraphic position directly overlying non-marine rocks with minor amounts of anhydrite and a lack of carbonates; (3) interbedded with non-marine red beds; (4) presence of borate minerals; and (5) association of subsequent igneous intrusions adjacent to the deposits. The Congo–Gabon basins in West Africa and the Sergipe Basin in east South America were formed during rifting and were possibly interconnected with non-marine rift lakes and hot hydrothermal CaCl2 -rich waters prior to continental rifting. Such rift setting is similar to modern-day lakes Magadi and Natron and they contain sodium carbonate brines (Eugster and Hardie, 1978). Deep and hot cal-

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

cium-rich brines of hydrothermal origin commonly flow through deep-seated faults such as at the bottom of the Red Sea (Craig, 1969). In modern settings where calcium magnesium-rich brines exists, tachyhydrite does not deposit even under dry conditions. Tachyhydrite would easily form if brine is extraordinarily hot, and under geothermal conditions which would then combine the source of heat and the chemistry of water. Boron-rich inflow, reflected in deposition of borate minerals, suggests either continental (from granitic terranes) or geothermal input into the Khorat Plateau. In summary, the Maha Sarakham saline giant was deposited in an extensive largely land-locked, but partly marine, evaporite basin in southeast Asia during the Cretaceous. This basin received influx from the world ocean through an inlet at the present southwestern corner of what is now the Khorat Plateau. Marine inflow was apparently choked off or ceased during the formation of the red beds that lie between the three salt phases. The basin experienced a substantial input of marine water during three major episodes that most likely represent the product of short-term sea-level rises. The apparent excess amounts of calcium, present in the deposits of tachyhydrite, probably originated from the replacement of gypsum by halite. Potash mineralisation resulted from extreme evaporation of seawater under very arid conditions; however, some non-marine input into the saline environment took place, based on the presence of borates.

8. Summary and conclusions The Khorat Plateau area of southeast Asia underwent periods of marine influx due to relative sea-level rise but was sporadically isolated from the world ocean. These evaporites mark a major marine deposition over the Khorat Plateau in northeast Thailand, following a long depositional history of non-marine sedimentation of the Khorat Group. Evaporites of the Maha Sarakham Formation were deposited as a single evaporite giant during the Cretaceous to early Tertiary in the Khorat and the Sakon Nakhon sub-basins. These two sub-basins are separated by the northwest-trending Phu Phan anticline which was formed during the Tertiary collision of

59

southeast Asia and south China. The stratigraphy and depositional features of the formation are similar throughout the plateau area. The formation is composed of three depositional members which include three evaporitic successions, separated by nonmarine clastic red beds that include some anhydrite nodules and beds. The formation thickens towards the basin centre in both basins suggesting basin subsidence during sedimentation. Evaporites include halite, anhydrite and a considerable accumulation of potassic minerals (sylvite and carnallite). Some tachyhydrite and minor amounts of borates are also present in the potash section suggesting input of non-marine or hydrothermal waters into the Khorat Plateau during deposition of evaporites. Sedimentary facies and textures of both halite and anhydrite suggest deposition in a shallow salinepan environment. Interbedded with the halite beds are anhydrite outlines of gypsum precursor crystals that form and which are the product of early diagenetic replacement by halite of primary shallow-water gypsum in saline ponds under relatively elevated temperatures caused by solar heating of the saline pan. The sulphur isotopic values of anhydrite interbedded with halite range from 14.3‰ to 17.0‰ (CDT), suggesting a marine origin for this sulphate. Bromine concentration in all halite member begins around 70– 200 ppm and systematically increases upward to 400 ppm towards the top of the halite, also suggesting a marine water source for the Maha Sarakham salts. Tachyhydrite found with the potassic salts resulted from releasing of calcium into the restricted waters of the basin due to the replacement of gypsum by halite replacement of gypsum, and early hydrothermal input of calcium chloride-rich waters into the depositional area. Erosion and influx of water from surrounding Triassic-age granitic terrains is the possible source for borates associated with potash-rich beds; however, hydrothermal influx is also possible. The non-marine red beds interbedded with the evaporites are fluvial or alluvial deposits and include displacive anhydrite nodules and beds and displacive halite in cubic forms. The low δ34 S isotopic values of this anhydrite suggesting non-marine sources for sulphate.

60

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Acknowledgements We are grateful to the Deputy Director General of the DMR in Bangkok (Department of Mineral Resources) of Thailand Mr. Prayong Augsuwathana for permission to study and sample the Maha Sarakham cores and to Mr. Thawat Japakasetr of the DMR for his encouragement and for permitting the study. Thanks to Mr. Nares Sattayarak of the Petroleum Exploration Group in the DMR for sharing ideas with us about the geology of the Khorat Plateau area and for his encouragement. The DMR has kindly funded some of the needed field expenses in northeastern Thailand. We thank the personnel of the Ground Water Division of the DMR in Khon Kaen for their support and for allowing us to use their core facilities. Gratitude is extended to Bruce Sellwood and Sedimentary Geology reviewers M.M. Blanc-Valleron and Peter Sonnenfeld for their valuable suggestions. This research was funded by the ARC-2445 grant (Australian Research Council) supporting M.E. as an ARC Postdoctoral Research Fellow.

References Achache, J., Courtillot, V., Besse, J., 1983. Paleomagnetic constraints on the late Cretaceous and Cenozoic tectonics of southeastern Asia. Earth Planet. Sci. Lett. 63, 123–136. Braitsch, O., 1971. Salt Deposits — their Origin and Composition. Springer, Berlin, 297 pp. Bryant, R.G., Sellwood, B.W., Millington, A.C., Drake, N.A., 1994. Marine-like potash evaporite formation on a continental playa: case study from Chott el Djerid, southern Tunisia. Sediment. Geol. 90, 269–291. Bunopas, S., Vella, P., 1992. Geotectonics and Geologic Evolution of Thailand. In: Chroen Pianchroen (Ed.), National Conference on Geologic Resources of Thailand: Potential for Future Development, pp. 209–228. Burke, K., 1975. Atlantic evaporites formed by evaporation of water spilled from Pacific, Tethyan, and Southern oceans. Geology 3, 613–616. Busson, G., Schreiber, B.C., 1997. Sedimentary Deposition in Rift and Foreland Basins in France and Spain (Paleogene and Lower Neogene). Columbia University Press, New York, 479 pp. Butler, G.P., Harris, P.M., Kendall, C.G.St.C., 1983. Recent evaporites from the Abu Dhabi coastal flats. In: Handford, C.R., Loucks, R.G., Davies, G.R. (Eds.), Depositional and Diagenetic Stectra of Evaporites — A Core Workshop. Soc. Econ. Paleontol. Mineral., Core Workshop 3, 33–64. Casas, E., Lowenstein, T.K., Spencer, R.J., Pengxi, Z., 1991.

Carnallite mineralization in the nonmarine, Qaidam Basin, China: evidence for the early diagenetic origin of potash evaporites. J. Sediment. Petrol. 62, 881–898. Cendon, D.I., Ayora, C., Pueyo, J., 1998. The origin of barren bodies in the Subiza potash deposit, Navarra, Spain: implications for sylvite formation. J. Sediment. Res. 68, 43–52. Cercone, K.R., 1988. Evaporative sea-level drawdown in the Silurian Michigan Basin. Geology 16, 387–390. Clabaugh, P.S., 1962. Petrofabric study of deformed salt. Science 136, 389–391. Claypool, G.E., Holser, W.T., Kaplan, I.R., Sakai, H., Zak, I., 1980. The age curves of sulphur and oxygen isotopes in marine sulfate and their mutual interpretation. Chem. Geol. 28, 199–260. Cooper, M.A., Herbert, R., Hill, G.S., 1989. The structural evolution of Triassic intermontane basins in northeastern Thailand. In: Thansuthipitak, T. (Ed.), Proceedings International Symposium on Intermontane Basins: Geology and Resources, Chiang Mai, pp. 231–242, Craig, H., 1969. Geochemistry and origin of the Red Sea Brines. In: Degens, E.J., Ross, D.A. (Eds.), Hot Brines and Recent Heavy Metal Deposits in the Red Sea. Springer, New York, pp. 208–242. deRuiter, P.A., 1979. The Gabon and Congo basins’ salt deposits. Econ. Geol. 74, 419–431. Drumm, A., Heggemann, H., Helmcke, D., 1993. Contribution to the sedimentology and sedimentary petrology of the non-marine Mesozoic sediments in northern Thailand (Phrae and Nan Provinces). Int. Symp. Biostratigraphy of Mainland Southeast Asia: Facies and Paleontology, 31 January–5 February, Chiang Mai, pp. 299–318. El Tabakh, M., Schreiber, B.C., Utha-Aroon, C., Coshell, L., Warren, J.K., 1998. Diagenetic origin of Basal Anhydrite in the Cretaceous Maha Sarakham salt: Khorat Plateau, NE Thailand. Sedimentology 45. Eugster, H.P., Hardie, L.A., 1978. Saline lakes. In: Lerman, A. (Ed.), Lakes, Chemistry–Geology–Physics. Springer, New York, pp. 237–293. Felmy, A., Wear, J.H., 1986. The prediction of borate mineral equilibria in natural water: application to Searles Lake, California. Geochim. Cosmochim. Acta 50, 2771–2783. Gardner, L.S., Haworth, H.F., Na Chiengmai, P., 1967. Salt resources of Thailand. Rep. Invest. DMR, Bangkok, 100 pp. Geisler-Cussey, D., 1982–1984. De la mer au sel: les facie`s superficiels de marais salants de Salin-de-Giraud (Sud de la France). Ge´ol. Me´dit. Marseille (IV) 9 (4), 521–549. Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctuating sea level since the Triassic. Science 235, 1156–1167. Hardie, L., 1990. The roles of rifting and hydrothermal CaCl2 brines in the origin of potash evaporites: a hypothesis. Am. J. Sci. 290, 43–106. Harvie, C.E., Eugster, H.P., Wear, J.H., 1982. Mineral equilibria in the six component seawater systems: Na–K–Mg–Ca–SO4 – Cl–H2 0 at 25ºC, II. Composition of the saturated solutions. Geochim. Cosmochim. Acta 46, 1603–1607. Hite, R.J., 1971. Potential for potash and related mineral re-

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62 sources, Khorat Plateau, northeast Thailand and central Laos. Unpubl. U.S. Geol. Surv. Proj. Rep. (IR) EA-1, 64 pp. Hite, R.J., 1974. Evaporite deposits of the Khorat Plateau, northeastern Thailand. Northern Ohio Geological Society, Proc. Fourth Symposium on Salt, pp. 135–146. Hite, R.J., Japakasetr, T., 1979. Potash deposits of Khorat Plateau, Thailand and Laos. Econ. Geol. 74, 448–458. Holser, W.T., 1966. Diagenetic polyhalite in Recent salt from Baja California. Am. Mineral. 51, 99–109. Holser, W.T., 1979. Mineralogy of evaporites. In: Burns, R.G. (Ed.), Marine Minerals. Reviews in Mineralogy 6, Mineralogical Society of America, Washington, DC, pp. 235–295. Holt, B.D., Engelkemeir, A.G., 1970. Thermal decomposition of barium sulfate to sulphur dioxide for mass spectrometric analysis. Anal. Chem. 42, 1451–1453. Holwerda, J.G., Hutchinson, R.W., 1968. Potash-bearing evaporites in the Danakil area, Ethiopia. Econ. Geol. 63, 124– 150. Hovorka, S.D., 1992. Halite pseudomorphs after gypsum in bedded anhydrite — clue to gypsum–anhydrite relationships. J. Sediment. Petrol. 62, 1098–1111. Hsu¨, K.J., 1972. Origin of saline giants: a critical review after the discovery of the Mediterranean evaporite. Earth Sci. Rev. 8, 371–396. Hutchison, C.S., 1989. Geological Evolution of South-East Asia. Clarendon Press, Oxford. Jowett, E.C., Cathles Jr., L.M., Davis, B.W., 1993. Predicting depths of gypsum dehydration in evaporitic sedimentary basins. Bull. Am. Assoc. Pet. Geol. 77, 402–413. Kendall, A.C., 1989. Brine mixing in the Middle Devonian of western Canada and its possible significance to regional dolomitization. Sediment. Geol. 64, 271–285. Kinsman, D.J.J., 1976. Evaporites: relative humidity control of primary mineral facies. J. Sediment. Petrol. 46, 273–279. Kyle, J.R., 1991. Evaporites, evaporitic processes and mineral resources. In: Melvin, J.L. (Ed.), Evaporites, Petroleum and Mineral Resources. Elsevier, Amsterdam, pp. 477–534. La Moreaux, P.E., Charal-Javanaphet, C., Jalichan, N., Na Chiangmai, P., Bunnang, D., Thavisri, A., Rakprathum, C., 1959. Reconnaissance of the geology and groundwater of the Khorat Plateau, Thailand. USGS Water Supply Pap. 1, 429 pp. Larsen, J.G., 1985. In: Schreiber, B.C., Homer, H.L. (Eds.), Sixth Int. Symp. on Salt 1, 323–330. Lowenstein, T.K., Hardie, L.A., 1985. Criteria for the recognition of salt-pan evaporites. Sedimentology 32, 627–644. Lowenstein, T.K., Spencer, R.J., 1990. Syndepositional origin of potash evaporites: petrographic and fluid inclusion evidence. Am. J. Sci. 290, 1–42. Lowenstein, T.K., Spencer, R.J., Pengxi, Z., 1989. Origin of ancient potash evaporites: clues from the modern nonmarine Qaidam Basin of Western China. Science 245, 1090–1092. Mouret, C., 1994. Geological history of northeastern Thailand since the Carboniferous: relations with Indochina and Carboniferous to early Cenozoic evolution model. Proc. Int. Symp. Stratigraphic Correlation of Southeast Asia, Bangkok, pp. 132–158. Nurmi, R.D., Friedman, G.M., 1977. Sedimentology and deposi-

61

tional environments of basin-central evaporites, Lower Salina Group (Upper Silurian), Michigan Basin. In: Fisher, J.H. (Ed.), Reefs and Evaporites — Concepts and Depositional Models. Am. Assoc. Pet. Geol., Stud. Geol. 5, 23–52. Ortı´, F., 1997. Continental evaporitic sedimentation on the South East border of the Ebro Basin (Catala´nides) during the Paleocene to Lower Oligocene. In: Busson, G., Schreiber, B.C. (Eds.), Sedimentary Deposition in Rift and Foreland Basins in France and Spain (Paleogene and Lower Neogene). Columbia University Press, New York, pp. 387–396. Ortiz, L.R., Mur, J.P., 1984. Geochemical characteristics of the potassium salt formation of Navarra (Upper Eocene); comparison with the Catalan potassic basin. Acta Geol. Hisp. 19, 81–95. Racey, A., Goodall, J.G.S., Love, M.A., Polachan, S., Jones, P.D., 1994. New age data for the Mesozoic Khorat group of northeast Thailand. Proc. Int. Symp. Stratigraphic Correlation of Southeast Asia, Bangkok, pp. 245–252. Richter-Bernburg, G., 1972. Sedimentological problems in saline deposits. UNESCO, Geology of Saline Deposits, pp. 33–37. Sattayarak, N., Polachan, S., Charusirisawad, R., 1991. Cretaceous rock salt in the northeastern part of Thailand. Proc. 7th Reg. Conf. Geology and Mineral Resources Southeast Asia (GEOSEA VII), Bangkok (Abstr.). Schreiber, B.C., 1988. Evaporites and Hydrocarbons. Columbia University Press, New York, 475 pp. Schreiber, B.C., Kinsman, D.J.J., 1975. New observations on the Pleistocene evaporites of Montallegro Sicily and a modern analog. J. Sediment. Petrol. 45, 469–479. Schreiber, B.C., Walker, D., 1992. Halite pseudomorphs after gypsum: a suggested mechanism. J. Sediment. Petrol. 62, 61– 70. Schreiber, B.C., Roth, M.S., Helman, M.L., 1982. Recognition of primary facies characteristics of evaporites and the differentiation of these forms from diagenetic overprints. In: Handford, C.R., Loucks, R.D., Davies, G.R. (Eds.), Depositional and Diagenetic Spectra of Evaporites — a Core Workshop. Soc. Econ. Paleontol. Mineral., Core Workshop 3, 1–32. Sengor, C.A.M., 1979. Mid-Mesozoic closure of Permo-Triassic Tethys and its implications. Nature 279, 590–593. Sengor, A.M.C., 1984. The Cimmeride orogenic system and the tectonics of Eurasia. Geol. Soc. Am. Spec. Pap. 195, 82 pp. Shearman, D.J., 1970. Recent halite rock, Baja California, Mexico. Inst. Mining Metall. Trans. 79B, 155–162. Shearman, D.J., 1978. Halite in sabkha environments. In: Dean, W.E., Schreiber, B.C. (Eds.), Marine Evaporites. Soc. Econ. Paleontol. Mineral., Short Course 4, 30–42. Sloss, L., 1969. Evaporite deposition from layered solutions. Bull. Am. Assoc. Pet. Geol. 53, 776–789. Smith, P.F.L., Stokes, R.B., Bristow, C., Carter, A., 1996. Mid-Cretaceous inversion in the northern Khorat Plateau of Laos PDR and Thailand. In: Hall, R., Blundell, D.J. (Eds.), Tectonic Evolution of Southeast Asia. Geol. Soc. London Spec. Publ. 106, 233–246. Sonnenfeld, P., 1984. Brines and Evaporites. Academic Press, New York, 613 pp.

62

M. El Tabakh et al. / Sedimentary Geology 123 (1999) 31–62

Sonnenfeld, P., 1995. The color of rock salt — a review. Sediment. Geol. 94, 267–276. Spencer, R.J., 1983. The role of pore fluids in evaporitic basins. Sixth International Symposium on Salt 1, 231–245. Utha-Aroon, C., 1993. Continental origin of the Maha Sarakham evaporites, northeastern Thailand. J. Southeast Asia Earth Sci. 8, 193–203. Veigas, J.G., 1997. First continental evaporitic phase in the south Pyrenean Central area: tremp gypsum. In: Busson, G., Schreiber, B.C. (Eds.), Sedimentary Deposition in Rift and Foreland Basins in France and Spain (Paleogene and Lower Neogene). Columbia University Press, New York, pp. 335– 342. Wardlaw, N.C., Nicholls, G.D., 1972. Cretaceous evaporites of

Brazil and West Africa and their bearing on the theory of continental separation. 24th International Geological Congress, Montreal, Sect. 6, pp. 43–55. Warren, J.R., 1982. The hydrological setting, occurrence and significance of gypsum in late Quaternary salt lakes in South Australia. Sedimentology 29, 609–638. Williams, G.E., 1991. Milankovitch-band cyclicity in bedded halite deposits contemporaneous with late Ordovician–Early Silurian glaciation, Canning Basin, Western Australia. Earth Planet. Sci. Lett. 103, 143–155. Williams-Stroud, S., 1994. The evolution of an inland sea of marine origin to a non-marine saline lake: the Pennsylvanian Paradox salt. Soc. Econ. Paleontol. Mineral. Spec. Publ. 50, 293–306.