Pleistocene calcretes from eastern Tunisia: The stratigraphy, the microstructure and the environmental significance

Journal of African Earth Sciences 58 (2010) 445–456

Contents lists available at ScienceDirect

Journal of African Earth Sciences

journal homepage: www.elsevier .com/locate / ja f rearsc i

Pleistocene calcretes from eastern Tunisia: The stratigraphy, the microstructureand the environmental significance

Wissem Gallala a,c,*, Mohamed Essghaïer Gaied b,c, Elhoucine Essefi d,e, Mabrouk Montacer b,c

a Département des Sciences de la Terre, Faculté des Sciences de Gabès, Cité Riadh, Zirig, 6072 Gabès, Tunisiab Faculté des Sciences de Sfax (FSS), Route de Soukra, Km 3.5, BP 802, 3018 Sfax, Tunisie Département des Sciences de la Terre, Tunisiac UR: GEOGLOB (Code 03/UR/10-02), Tunisiad École Nationale des Ingénieurs de Sfax (ENIS), Tunisiae UR: Dynamique Sédimentaire et Environnement (UDSE) (Code 03/UR/10-03), Tunisia

a r t i c l e i n f o a b s t r a c t

Article history:Received 26 February 2009Received in revised form 21 April 2010Accepted 23 April 2010Available online 4 May 2010

Keywords:PleistoceneCalcreteGeochemistryMineralogyGenesisEastern Tunisia

1464-343X/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.jafrearsci.2010.04.009

* Corresponding author at: Faculté des Sciences deKm 3.5, BP 802, 3018 Sfax, Tunisie Département desTel.: +21622874565; fax: +21674274437.

E-mail addresses: [email protected] (W. G(M.E. Gaied).

This paper is meant to study the stratigraphy, the mineralogy, the microstructure and the geochemistryof Pleistocene calcretes from eastern Tunisia in order to infer the environmental factors intervening intheir formation.

Samples of eight profiles of Pleistocene calcretes from eastern Tunisia were examined on the basis of avariety of techniques including Optical Microscopy (OM), Scanning Electron Microscopy (SEM), X-ray Dif-fraction (XRD), chemical analysis and Atomic Absorption Spectrophotometer (AAS) techniques. Then, theobtained data underwent a statistical analysis on the basis of Factor Analysis (FA) and Principal Compo-nent Analysis (PCA).

On the basis of field missions, five different horizons have been differentiated from bottom to top of allprofiles: nodular, powdery, massive Brecciated and laminar horizon. The mineralogical study shows twominerals categories inversely proportional: calcite and (quartz and the clay). It shows also shows thatPalygorskite is the dominant clay mineral. The escarpment edge is capped by a limestone containingfibrous palygorskite. Finally, superficial calcrete are described: a brecciated horizon which occurs inpockets on the plateau surface. This study about eastern Tunisia revealed the occurrence of successivecycles of calcretisation. Pedogenesis, water table oscillation, sedimentogenesis and stromatogenesis arethe intervening factors in the calcretisation process. During the Pleistocene, they interfered with eachother according to the climatic pulsations. From the studied case, it may be noticed that the formationof each calcrete horizon is the result of a dominating process that takes place during a distinguishablestage. In the first stage, the pedogenic process is developed by palygorskite formation including authi-genic replacement or formation from a precursor mineral, neoformation from the breakdown productsof such minerals or neoformation from suitable solutions. In the second stage, the powdery horizon isformed in the slope of the distal zone which presents a drained environment. In the third stage, severaldiagenetic processes (cementing, compaction, dissolution...) contribute to the formation of the laminarand massive horizon. Since it is exposed to dryness for a long period, the massive horizon is harderand more compact. In the fourth stage, the banding of light–dark in the laminar horizons reflects adry-wet season alternation seasons. Dark beds are formed by the stromatolitic cover were developed dur-ing the wet season, whereas light beds were developed in an extremely arid climate argued by the pres-ence of the detrital grains. In the fifth stage, the brecciated horizon, which occupies the channels, isformed by well rolled concretions, which present a dismantling material of Early and Middle Pleistocenecalcretes after the Post-Villafranchian compressive phase. Thus, calcretisation seems to have been con-trolled by periods of uplift and stability of the slope, given that calcrete formation might be inhibitedby the activation of the sedimentation of colluvial materials as a consequence of the tectonic activity.We also suggest that groundwater and biological activity may play a significant role in the developmentof pedogenic, sedimentological and polygenetic calcrete cycles within the same sedimentary basin. Thealternation of dry and wet climatic periods may be responsible for the calcrete genesis.

� 2010 Elsevier Ltd. All rights reserved.

ll rights reserved.

Sfax (FSS), Route de Soukra,Sciences de la Terre, Tunisia.

allala), [email protected]

1. Introduction

From a climatic point of view, calcretes are typical surface for-mations of the semi-arid and subhumid landscapes (Estrela and

446 W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456

Vogt, 1989). They extend from the desert regions of the Saharawith less than 100 mm rainfall per year to the subhumid Mediter-ranean zones with a rainfall ranging from 300 mm per year to500 mm per year (Wright and Tucker, 1991). It is worth noting alsothat calcretes bearing the Quaternary continental formations areubiquitous features in the Maghreb (Ballais and Ben Ouezdou,1991). From a paleontological point of view, these calcretes ofthe Maghreb share the same fauna; since the abundant terrestrialgastropods in eastern Tunisia are like those of Djebel Bargou(Northern Western Tunisia), Oranie (Northern Western Algeria)and near Taforalt, NW Morocco (Ducloux and Laouina, 1989). Asfor the Tunisian context, calcretes cover large areas in Tunisia,mainly in the eastern plain of the Sahel; they are commonly foundon Quaternary alluvial fans, terraces and glacis (Fig. 1). Chronolog-ically speaking, horizons of carbonate accumulations have beendeveloped in many periods during the Pleistocene (Kamoun,1981; Regaya, 2000). They are observed on glacis truncating differ-ent rocks with varied natures and ages. In eastern Tunisia, somehills (Djebel Bou Thadi, Ennagguer, Sidi Bou Ali. . .) are composedof thick continental detrital Quaternary formations. These depositsoften show a carbonated development leading to thick calcretebeds. Consequently, this study is meant to determine the distribu-

Fig. 1. Location of the studied calcrete profiles and geologic

tion of calcretes, their micromorphological, mineralogical andchemical characteristics and to suggest possible developmentpathways in eastern Tunisia.

2. Geological and structural setting

Tunisian eastern zones are characterized by a varied but notwell contrasted topography (Ben Jmaa, 2008); eventhough theyare interrupted by local uplands such us of Bou Thadi (240 m)and of El Jem (176 m), these areas make up the so-called Sahelplain. Hydrologically speaking, this plain is divided into smallerdrainage basins, which drain towards the sebkhas of the Sahel area(Fig. 1). For instance, the sebkha of Kelbia in the North East ofKairouan represents the hydrological outlet of the surrounding up-lands (Castany, 1947). Northward the uplands of Bou Thadi-ElJem-Jemmel, El Rhena and southward the upland of El Jem, thesebkhas of Mechertate, Chrita and Sidi El Hani make up the so-called the endorheic system of Mechetate-Chrita-Sidi El Hani(Ben Jmaa, 2008). From a hydrogeological point of view, in the Sa-hel area, many aquifers such as of Kairouan, Souassi and Zarmdinehave a connection with the surface; they converge towards the

al map of the Tunisian Sahel (Ben Haj Ali et al., 1985).

W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456 447

low lands namely towards the sebkhas surfaces; this groundwaterconvergence makes the water table next to the surface (Essefi,2009). Consequently, the groundwater contribution should be ta-ken into account when studying the calcretisation processes. Asfor the geodynamics context, the eastern Tunisian areas showNE–SW folds, in which continental and marine Miocene series ofclay and sand outcrop (Bédir, 1988). These series are conformableto the subjacent sediments. It is worth to be noted that marine Pli-ocene deposits, which outcrop on both sides of the anticlinal struc-tures, are slightly unconformable on the preceding Miocenecontinental series. Then, the Pliocene outcrops pass gradually tothe Villafranchian continental formations (Kamoun, 1981). Thislatter unit was folded before the Tyrrhenian transgression. Accord-ing to Bédir (1985, 1988) and Winnock and Bea (1979), the isopachcontours of the Plio-Quaternary sequences show the configurationof basins in the form of gutters or grabens, which follow E–W andNS tectonic corridors. These grabens limit a platform basin withreduced subsidence, but this subsidence covers a considerable re-gion and occurred during the period stretching from the Mesozoicto the Quaternary (Patriat et al., 2003). These geological structuresare mainly overlaid by quaternary and Holocene sandstones, con-tinental red bed and calcrete (Kamoun, 1981).

3. Methodology

To elaborate this study about calcretes of eastern Tunisia, dif-ferent field missions were carried out. During these missions,eight geological profiles were surveyed and sampled (Fig. 2). Col-lected samples underwent mineralogical, microscopic and geo-chemical studies. First, the mineralogical study is based on theX-ray Diffraction (XRD) to pinpoint changes in calcretes fromBou Thadi profile, namely percentages of quartz, clay and calcite.From this same profile, a sample (Fig. 2; BH1) was chosen to un-dergo a mineralogical study of its oriented clay powders afterheating at 500 �C for 2 h, after air-drying at 25 �C and after add-ing glycol solvated. Second, observations with Optical Microscope(OM) and Scanning Electron Microscopes (SEM) were useful tostudy the microstructure of calcrete samples from different pro-files. Third, the geochemical analysis of samples from all thestudied horizons was done on the basis of the Atomic Absorption

Fig. 2. Sections of selected calcrete profiles showing the arran

Spectrophotometer (AAS) techniques. The obtained data werestatistically studied on the basis of the Factor Analysis (FA)method.

4. Results

First, the different outcropping horizons representing Pleisto-cene calcretes of eastern Tunisia were identified. Then, the miner-alogical composition was followed along profiles. As a third move,the microstructures of samples were investigated to pinpoint theirstriking features. Eventually, the geochemical composition wasinvestigated on the basis of the ternary diagram (SiO2 + Al2O3 + -Fe2O3, CaO, MgO) and the Factor Analysis.

4.1. Calcretes occurrence

In order to do lithostratigraphic correlations and to study thegenetic evolution of the Pleistocene calcretes, a series of litho-stratigraphic profiles was surveyed in eastern Tunisia accordingto a North–South section, namely in the following sites: SidiBou Ali, Kalaâ Kébira, Kalaâ Sghira (Ennaguer), Menzel Hayet,Hencha, Jebiniana, Bou Thadi and El Ghraba (Fig. 1). When exam-ining the eight outcropping profiles (Fig. 2), the Quaternarydeposits in the studied areas share four similar characteristicsmaking them quite correlative. First, they occupy the tops andthe residuals of the hills. Second, the base of the calcretes consistsof a friable facies. Third, the lithification effect becomes morespectacular at the top. Fourth, the gradient of carbonate accumu-lation shows an upward enrichment. Sedimentologically speaking,five facies were found out during field missions; each facies occu-pies a horizon detectable by the naked eye in the field. It is worthto be noted that it is not of necessity for a horizon to be presentin all profiles; that is to say, except the ever present powdery,massive and laminar horizons, the others two horizons mayeclipse in some profiles. In the studied case, we have some pro-files containing all the horizons. For instance, the profile of BouThadi (Fig. 2) contains all horizons. From the base upward, wedistinguish the nodular, the powdery, the massive, the brecciatedand the laminar horizons.

gement of the different horizons. For location see Fig. 1.

Fig. 3. (A) Ennaguer profile Face (a and b: sand and clay intercalation; c: red loam bed with carbonated nodules; d: nodular horizon, e: powdery horizon; f: massive horizon).(B) Nodular horizon developed on muddy sandy red bed. (Hammer is 30 cm long.) (C) Composite elements resulting from reworking: previous components are embedded inbuff matrix with loose pisoliths and oolites. (D) Massive horizon rich in Helixes (Hammer is 30 cm long). (E) Presence of the channel deposits on a strongly folded massivehorizon. (Hammer is 30 cm long.) (F) Brecciated horizon made up of reworked anterior calcretes materials. (G) Middle Pleistocene powdery horizon rests directly on theblocks of Early Pleistocene massive horizon. (Hammer is 30 cm long.) (H) Prolonged subaerial exposure with widespread development of brecciation on massive horizon.

448 W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456

4.1.1. Nodular horizonIn the Sahel area, the thickness of this nodular horizon varies

between 1 m and 9 m (Burollet, 1956). In the studied sites, itsthickness varies from approximately 0.3 m in Kalaa Kebira toapproximately 2.5 m in Bou Thadi (Fig. 2). In the profile of KalaaSghira (Fig. 3 A, d), this horizon rests on a red1 loam bed withcarbonated nodules. Compared to the other horizons, this nodularfacies is relatively rich in clay. Moreover, this horizon shows an

1 For interpretation of color in Figs. 2, 3, 5, 7, and 9, the reader is referred to the webversion of this article.

extensive formation of clay minerals especially palygorskite. Asfor the nodules contained in this horizon, their colours vary fromchamois to white and their sizes range from 1 to 3 cm (Fig. 3 A,d; B); this variation should by rights reflect an increase of the car-bonated fraction in this horizon. Their morphology also variesalong a bottom-top axis from spherical to subspherical to irregularmorphology.

4.1.2. The powdery horizonAccording to the studied site, its thickness varies from approx-

imately 0.3 m in Hencha to approximately 0.8 m in Jebiniana

Fig. 4. Evolution of the mineralogical composition of Early Pleistocene calcretesalong the profile of Bou Thadi.

Fig. 5. Oriented clay powders; a: after heating at 500 �C for 2 h (sample BH1); b:air-dried at 25 �C; c: after adding glycol solvated, P: palygorskite, S: smectite, K:kaolinite, I: interstratified clay.

W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456 449

(Fig. 2). In the profile of Kalaa Sghira (Fig. 3 A, e), its thickness var-ies from 0.8 m to 1.5 m. The carbonated nodules and mycelium inthis dusty and soft facies are quite abundant; their length rangesbetween few millimetres to few centimetres. The matrix contain-ing these elements is characterized by a frequent occurrence ofplant root casts.

4.1.3. The massive horizonAccording to the studied site, the thickness of such a horizon

ranges between approximately 0.1 m in Sidi Bou Ali and approxi-mately 1.5 m in Jebiniana (Fig. 2). In the profile of Kalaa Sghira(Fig. 3 A, f), it occupies the top of the profile. The most strikingcharacteristics of this facies are its red salmon colour, its high me-chanic resistance and the frequent presence of Helixes. Added tothese features, this horizon generally shows the following distinc-tive criteria:

The presence of the ooids and pisoliths (Fig. 3 C).The tests of helixes are well preserved (Fig. 3 D).This calcrete is highly compact with shrinkage cracks on thewell ventilated surface (Fig. 3 H).The salmon colour is primarily due to the iron oxide generallyfixed in clays or in granules.The presence of excessive detrital materials and the presence ofchannels at its top (Fig. 3 H).

4.1.4. Brecciated horizonAccording to the studied site, the thickness of such a horizon

ranges between approximately 0 m in Kalaa Sghira and approxi-mately 0.4 m in Bou Thadi. To be more succinct, this brecciatedhorizon was observed only in the El Guellela and Bou Thadi zones.It is made up of centimetric and well rolled size calcrete remains(Fig. 3 F). From the base to the top, nodules become coalescent.The reworking of this material described in the Middle and LatePleistocene calcretes presuppose the intervention of diffused orchenalized streamings; such features were actually observed dur-ing field missions (Fig. 3 E).

4.1.5. Laminar horizonThe laminar horizon is a hard and thin calcareous accumulation

with sublithographic aspect. It underlies the surface of the massivehorizon. Along a vertical section, this horizon shows an alternationof clear and coloured layers. These thin millimetric layers areunderlined by ‘‘impurities” (organic matter, manganese, clay)(Durand et al., 1979).

4.2. Mineralogical characterization

The mineralogical study focused on the Early Plietocene cla-cretes of Bou Thadi profile (Fig. 4). The (XRD) shows that calcite,quartz, clays are the most significant minerals. Incidentally, thefeldspars appear with non significant amounts. As for their evolu-tion along the studied profile, the dominant minerals may be clas-sified into two categories inversely proportional. On the one hand,the calcite represents in itself a category; its amount generally in-creases from the bottom to the top of the profile. On the otherhand, the quartz and the clay represent the second category. Unlikethe calcite, their amounts generally decrease from the bottom tothe top of the profile (Fig. 4).

As for the mineralogical analyses of the clayey fraction (<2 lm)within a sample of calcrete from Bou Thadi profile (Fig. 2; BH1),they show that palygorskite, kaolinite, illite and smectite are pres-ent with variable amounts within this calcrete (Fig. 5). As it is quiteproved in the literature (Vanden Heuven, 1966; Gardner, 1972;Nahon and Ruellan, 1975; Watts, 1980), results also show thatpalygorskite is by excellence the dominant clay in this calcrete;

since it forms the dominant part of the coated clays. However,the interpretation of the geological significance of palygorskite for-mation has always been a matter of a controversy between schol-ars. Millot et al. (1969) and Singer and Norrish (1974) attributed itsorigin to the neoformation. Millot et al. (1977) and Paquet (1983)suggested the direct alteration of the silicated minerals or of thereplacement of aluminous detrital elements like montmorillonite.Palygorskite was also considered as a marker of hot and aridclimate on continental masses (Millot, 1964; Bolle et al., 1999).Furthermore, Jamoussi et al., (2003) went so far to consider thePalygorskite as an indicator of the compressive movements.

450 W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456

4.3. Microstructure and microscopic features

The observations of thin sections of different calcretes facies bythe (OM) and by (SEM) reveals that most of horizons present somestriking features, which may be good indicators of the different pro-cesses operating in calcrete formation; such processes may be in-ferred by studying the cement, the replacement and epigenesisprocess, the coated grains, the Palygorskite and the biogenic features.

4.3.1. CementThe characteristic of the cement was studied within massive

calcretes and transition between massive and laminar horizons.

Fig. 6. (A) peloids in microsparitic matrix (sample KK4). (B) The epegenesis allows the taffects the quartz grains: quartz grains have microspar coronas and patches of dark micrAbundance of badly graded quartz giving a packstone structure and presence of rootlecomposite ooids showing varied nuclei and several stages of ooidisation (sample BH3). (Hmicrite, Ms: microspar.

Calcretes are generally cemented by a micritic or microsparitic cal-cite. These two different types of cement reflect different environ-mental conditions during cementation. The micritic cement isdeveloped during the replacement of the sediments with high ratesof evapotranspiration under wet conditions. The opaque microspa-ritic cement represents less replacement and contains a higher per-centage of original grains (Maizels, 1990b); its most strikingcharacteristic is the important frequency of ooids, which givesthe rock a ‘‘perlitic” aspect (Vogt, 1984). As for the studied case,observations with Optical Microscope (OM) of samples from themassive calcretes of different profiles (Fig. 2; BH3, JB3, HE4, MZ4,KK4) show this important frequency of ooids bathing within a

otal replacement of Helix fragment (sample JB3). (C) Very marked epigenisis whichite are preserved between grains (sample HE4). (D) Explicative illustration of C. (E)t pattern underlined by microspar (sample MZ4). (F) Detail of rootlet pattern. (G)) Detail of G. we note all photographs are in polarized light. Q: quartz, C: calcite, M:

W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456 451

microsparitic calcite (Fig. 6 A); these peloids with or without nu-cleus in this microsparitic matrix impose a texture varying fromgrainstone to packstone (Fig. 6 E). Even when the nucleus is pres-ent, it can be individualized as a single cortex with a great difficulty(Fig. 7A–C). It should be noted that these ooids are seldom asym-metrical and graded bedded. This dense structure with a tendencytoward an organization is a sign of a reworked material, which pro-vides with a mixture of detrital materials and sometimes withcomposite ooids testifying a polyphasage (Fig. 6G and H). Micrite

Fig. 7. (A) The quartz grain can form nuclei of a coated grains with a micritic cortex (samof A and B. (D) Stromatolithic mat showing the alternating bands of yellowish micrite anthe latter has a chollomorphe structure pointing out the shape of the algaire mat (samplconcentrated in fissure allotted to a eolian origin (sample MZ2).

and other elements can be affected by crack networks (Fig. 6 E),which are filled with the limpid microsparite resulting from thelate diagenesis (Fig. 6 F). Euhedral to sub-euhedral microspariteand sparite are both present in close proximity to host clasts to-wards the centre of void spaces. The abundance of these cracksor veins reflects the influence of withdrawal processes (Khadkikaret al.., 2000) which can be due to seasonal temperature variation.Moreover, the micromorphological analyses of the calcrete withScanning Electron Microscopy (SEM) present a polycyclic and

ple NG4). (B) Reworked peloids with fine quartz and micrite (sample BH3). (C) Detaild dark organic matter (sample JB3). (E) Transition Zone of massive-laminar horizon,e GR4). (F) Quartz present in the laminar horizon with reduced size disseminated or

452 W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456

polyphased development of calcite around quartz grains (Fig. 8 A)(micrite 1–4 lm, microsparite 4–20 lm, sparite higher than20 lm). On the other hand, observations with Optical Microscope(OM) show that laminar horizons are characterized by the alterna-tion of dark–light beds. Light beds (Fig. 7D and E) are made up ofdome-structured micrite or bearings. They primarily consist ofspheroid microsparite with a fibroradial growth. The dark bedsconsist of organic matter and of oxides. The quartz is also presentin a reduced size, concentrated or disseminated (Fig. 7 F). Analyseswith Scanning Electron Microscopy (SEM) show these grains ofquartz situated within a micritic matrix (Fig. 8 B); these grains

Fig. 8. (A) Developed rhombohedral calcite grain-coating micritic cement (sample BH3cement within a pore space (sample BH2 upper part). (D) Quartz and calcareous clasts ri(E) Well rouned quartz and carbonate clasts rimmed by micrite grain-coating cements wAggregation of particles coated by needle-fibre calcite (sample BH2 lower part). (G) SEMBH2 lower part). (H) Presence of fine fibrous palygorskite (red bed) (sample BH1).

may have an eolian origin (Coque, 1962; Vogt, 1984; Verrecchia,1994; Regaya, 2000, 2002) and can undergo corrosion phenomena.

4.3.2. Replacement and epigenesis processThis feature is studied within the massive calcrete from Hencha

profile (Fig. 2; HE4). The porosity is present under the form of dis-solution of allochems that will be replaced by microsparite. Animportant phenomenon that affects the quartz grains is the silicacorrosion and its replacement by calcite (Fig. 6C and D). Even-though this epigenisis affects the quartz grains by a surroundingmicrospar, patches of dark micrite are preserved between grains.

). (B) Quartz grain situated within a micritic matrix (sample BH3). (C) Needle-fibremmed by micrite grain-coating cements and palygorskite (sample BH2 upper part).ith localised pore-filling micrite cement development (sample BH2 lower part). (F)image showing a network of calcified filaments from the powdery horizon (sample

Table 1Table of chemical analyses in different calcrete beds.

LOI CaO MgO SiO2 Fe2O3 Al2O3 Na2O K2O

HE1 40.37 46.20 0.93 9.24 0.33 0.68 0.07 0.15HE2 40.94 47.04 0.91 8.12 0.26 0.45 0.19 0.20HE3 32.29 40.08 1.10 22.83 0.55 1.21 0.05 0.26MZ1 40.63 47.88 0.60 8.56 0.33 0.53 0.03 0.10MZ2 38.86 46.48 0.52 1172 0.74 1.13 0.05 0.32MZ3 40.88 48.72 0.80 8.13 0.30 0.45 0.03 0.08MZ4 40.98 46.48 0.70 8.98 0.29 0.45 0.07 0.11JB1 38.74 49.28 0.57 10.15 0.29 0.45 0.04 0.14JB2 22.25 28.56 1.16 42.79 0.49 1.74 1.08 0.43JB3 15.65 18.48 0.99 56.48 0.96 2.64 0.28 0.68KK1 18.19 19.60 1.00 54.33 1.80 4.16 0.11 0.76KK2 34.33 40.04 1.06 21.39 0.48 0.76 0.20 0.22KK3 41.77 48.44 0.70 7.02 0.20 0.38 0.05 0.08KK4 41.52 47.32 0.62 7.27 0.27 0.53 0.09 0.15SB1 18.56 21.56 0.89 52.20 1.58 3.63 0.16 0.66SB2 25.87 29.70 1.06 38.51 1.03 2.42 0.1 0.44SB3 36.72 41.40 0.70 17.11 0.43 0.91 0.05 0.19NG4 21.13 25.36 0.67 38.34 1.06 2.42 0.13 0.63NG3 35.17 39.40 0.99 20.11 0.67 1.51 0.04 0.27NG2 43.10 51.60 0.63 3.00 0.14 0.23 0.05 0.05NG1 42.90 52.44 0.73 1.72 0.15 0.23 0.05 0.06BH1 40.68 46.80 0.97 8.56 0.27 0.45 0.05 0.12BH2 33.58 39.20 1.35 20.11 0.64 1.21 0.04 0.22BH3 25.09 27.16 1.99 38.51 1.37 3.48 0.23 0.55BH4 39.18 43.40 0.68 11.30 0.47 0.91 0.06 0.17GR1 39.84 45.40 0.65 9.58 0.40 0.91 0.06 0.17GR2 40.31 45.36 0.76 8.90 024 0.38 0.05 0.10GR3 26.53 29.40 1.35 38.51 0.54 1.36 0.19 0.33

Fig. 9. Ternary diagram sowing the distribution of major chemical elements incalcretes.

W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456 453

The interaction of water with the grains causes dissolution (Fig. 6);since the silica is not saturated in the pores (Wang et al., 1994).

4.3.3. Coated grainsThis feature is studied within samples from the powdery and

nodular horizon of Bou Thadi profile (Fig. 2; BH2). In the upper partof this horizon, some clasts exhibit mixed clay (palygorskite) andmicritic coatings (Fig. 8 D). In the lower part of this nodular hori-zon, grain-coating micritic cements are more clearly crystalline;they are formed around quartz and carbonate clasts (Fig. 8 E).The particle aggregations can be coated by needle-fibre calcite(Fig. 8 F).

4.3.4. PalygorskiteThe occurrence of palygorskite suggests a chemical sedimenta-

tion enriched in magnesium under a semi-arid climate (Bustillo etal., 2002). In the studied profiles, samples occasionally displaypalygorskite fibres occupying an intercrystalline microporosity(Fig. 8 C) and coating grains (Fig. 8 H) in the elongated fibrousred bed. Under Scanning Electron Microscopy (SEM) observation,the palygorskite was clearly noticed within the upper part of thePlio-Quaternary passage of Bou Thadi profile (Fig. 2; BH1).

4.3.5. Biogenic featuresAlveolar septal structures are commonly recognised in calcretes

and their formation is considered as the result of fungal activity insymbiotic association with roots (Wright, 1986). In the studiedprofiles, networks of calcified filaments are shown in the powderyhorizon (Fig. 2; BH2). Under SEM, the septa are seen as calcified fil-aments with 100 lm long and 2–5 lm diameter (Fig. 8 G). Thesecalcified filaments consist of a dense packing of rhombohedral-shaped crystals, 1 lm in size, probably of calcite. In some cases,some crystals grow perpendicularly to the filaments with a lengthexceeding 30 lm and a width reaching 5 lm (Fig. 8 G).

4.4. Geochemical study

Chemical analyses of different calcrete beds (Table 1) confirmthe conclusions of the mineralogical analyses. For instance, theSiO2, which is the major component of the quartz and the clays,is inversely proportional to the CaO, which is the major componentof the calcite. Moreover, Al2O3, TiO2, Fe2O3 are proportional to SiO2.Furthermore, the Ca/Mg ratio is high, indicating the predominanceof calcite in all calcretes. For instance, MgO percentages do not ex-ceed the 2%. Nevertheless, highest values are recorded for the nod-ulous beds, testifying the presence of palygorskite, which waspreviously identified by (XRD). As for the interpretation of paly-gorskite occurrence, the formation of such authigenous clays likepalygorskite in calcretes is the result of an enrichment of vadosewater with magnesium after magnesian and calcite precipitationin calcrete (Watts, 1980). Gauthier-Lafaye et al., (1993) announcedthat the isotopic study of palygorskite oxygen, enriched in 18O(d18O = + 33% to 34‰ SMOW), and calcite carbon oxygen associ-ated with calcretes from Portugal and Morocco highlights theimportance of the evaporation of the meteoric water in this min-eral formation. Moreover, these results show that, under a semi-arid climate, palygorskites crystallize during periods of a strongevaporation, whereas calcites crystallize during wet periods. Thisassumption is confirmed by Khadkikar et al. (2000) who suggestedthat, if calcretes are related to sepiolite/palygorskite, with the pres-ence of smectite and hematite, then the studied calcretes are morethan likely to be developed under a semi-arid climate (mean an-nual rainfall between 100 and 500 mm); such a suggestion is sim-ilar to the present climate of the studied area in eastern Tunisia.

To visualize the evolution of the geochemistry of different pro-files, the massive, chalky and powdery horizons are presented in

the ternary diagram (SiO2 + Al2O3 + Fe2O3, CaO, MgO) (Fig. 9). Thecommon feature between all samples is their divergence formthe magnesium pole. Eventhough samples of the three horizonsoverlap on the ternary diagram, samples of each horizon may makeup a standalone population. First, the powdery population is char-acterized by a high rate of (SiO2 + Al2O3 + Fe2O3) (between approx-imately 30% and 40%), as a result of the presence of the clayeyfraction. Second, the chalky population is characterized by a med-ium rate of (SiO2 + Al2O3 + Fe2O3) (between approximately 10% and30%), as a result of the presence of the clayey fraction in the discon-tinuous beds. Third, the massive population is characterized by alow rate of (SiO2 + Al2O3 + Fe2O3) (between approximately 2% and15%), as a result of the presence of the free quartz. It is worth men-tioning that this vertical differentiation on the ternary diagram(SiO2 + Al2O3 + Fe2O3, CaO, MgO) testifies the genesis by jerks alongthe studied profiles (Regaya, 2000).

Using the ‘‘Statview” package, data obtained from the labora-tory analysis were used as variable inputs for Factor Analysis. Table2 shows the Factor Analysis of the major element concentrations incalcrete. Results of the Principal Component Analysis (PCA) indi-cate that two or three factors explain most of the data variance.

Table 2Factor analysis of major element concentrations data.

Variables Factor 1 Factor 2 Factor 3

LOI �0.979 �0.035 �0.037CaO �0.981 �0.013 �0.078MgO 0.612 �0.434 �0.092SiO2 �0.02 0.897 0.039Fe2O3 �0.08 �0.164 0.961Al2O3 0.941 0.142 0.079Na2O 0.503 �0.24 0.229K2O 0.957 0.22 0.064Proportion of variance (%) 60.6 16.1 11.9Cumulative (%) 60.6 76.7 88.6

454 W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456

One element was chosen as a tracer for each factor. Factor 1 has ahigh loading of K2O, Al2O3 and MgO (60.6% of the total variance); ithas strong negative correlations with CaO. This factor can be as-cribed to the presence of the clay. Factor 2, which represents16.1% of the total variance, includes SiO2. This factor reflects thesignatures of quartz which consist of an important phase of cal-crete. Factor 3 represents 11.9% of the total variance of the dataand exhibits high factor loading of Fe2O3. This iron oxide is respon-sible for the salmon colour of calcretes.

5. Discussion

Results of the mineralogical, microscopic and geochemicalstudies are in a total agreement. Indeed, calcite and quartz aretwo dominant minerals within the majority of the encrustedbeds. When it is in the matrix, calcite is presented in the micriteform; and it is in the microsparite form in the cases of quartzepigenesis and the dissolution recrystallization. Pedogenesis,water table oscillation, sedimentogenesis and stromatogenesisare the intervening factors in the calcretisation process. Duringthe Pleistocene, they interfered with each other according tothe climatic pulsations. From the studied case, it may be noticed

Fig. 10. Schematic Diagram illustrating the suggested dep

that the formation of each calcrete horizon is the result of adominating process that takes place during a distinguishablestage.

In the first stage, the pedogenic process is developed by paly-gorskite formation including authigenic replacement or formationfrom a precursor mineral, neoformation from the breakdown prod-ucts of such minerals or neoformation from suitable solutions(Isphording, 1973; Singer and Norrish 1974). Such a model seemsto be supported in eastern Tunisia where palygorskite is developedwithin sediments deposited in an alkaline (saline) environmentunder a seasonally arid climate. Such aridity is indicated by thehigh amounts of magnesium and silica compared to the aluminiumamounts (Botha and Hughes, 1992). The pedogenesis, the microor-ganism activity, the evapotranspiration and the percolation lead tothe nodules and myceliums formation that is the result of CaCO3

precipitation around the plants roots during the dry season(Fig. 10). In vegetated areas, plants cause cracking in the surfacepart of the red bed. These cracks will increase and become filledwith water and particles. In this situation, the evapotranspirationallows the precipitation of CaCO3 from residual water coming bywater-table rise (Fig. 10).

In the second stage, the powdery horizon is formed in the slopeof the distal zone which presents a drained environment (Freytetand Plaziat, 1978). When the climatic environment enhanceddevelopment of vegetation, the chemical sedimentation is espe-cially dominant. The soluble migrating phase, in the sheet washarea, includes all the alkaline and calcano-earthy bases, as wellas the silica of silicates other than kaolinite and, in some condi-tions, iron and manganese. The solubilization of these elementsis facilitated by the abundance of the organic matter depositedon the surface and by the biological activity, which predominatesin these area. The residual phases include inalterable minerals suchas kaolinite, quartz, alumina hydroxide and iron hydroxides (Dur-and, 1956). The presence of faded gastropod tests suggested thatdepositing conditions were more than like occurring in an oxygen-ated environment.

osit model of different types of horizon development.

W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456 455

In the third stage, several diagenetic processes (cementing,compaction, dissolution...) contribute to the formation of the lam-inar and massive horizon. Since it is exposed to dryness for a longperiod, the massive horizon is harder and more compact. These cli-matic conditions allow the cracking of these calcretes which willbe filled with detrital elements during the wet season. The thick-ness of this type of calcrete is controlled by CO2 pressure, evapora-tion intensity and dry season duration (Wang et al., 1994). LowPCO2, intense evaporation and long dry seasons result in producingthicker calcretes and vice versa. Silica deposition would also occurwhere a silica-rich runoff or groundwater is locally concentrated.

In the fourth stage, the banding of light–dark in the laminarhorizons reflects a dry-wet season alternation seasons. Dark bedsare formed by the stromatolitic cover were developed during thewet season, whereas light beds were developed in an extremelyarid climate argued by the presence of the detrital grains; thesegrain may have an eolian origin (Fig. 10) and a low calcite precip-itation. The results obtained are similar to the literature data (Elloyand Thomas, 1981; Estrela and Vogt, 1989; Freyet and Verrecchia,1989; Verrecchia, 1994; Regaya, 2000, 2002) and allow us to pro-pose a biological origin for the laminar horizon and to clarify cer-tain points concerning its formation: this horizon often affectsthe summit part of carbonated encrusting. It shows many similar-ities to algal constructions of stromatolite type.

In the fifth stage, the brecciated horizon, which occupies thechannels, is formed by well rolled concretions, which present a dis-mantling material of Early and Middle Pleistocene calcretes(Regaya, 2000) after the Post-Villafranchian compressive phase(Kamoun, 1981; Bouaziz et al., 2002). Thus, calcretisation seemsto have been controlled by periods of uplift and stability of theslope, given that calcrete formation might be inhibited by the acti-vation of the sedimentation of colluvial materials as a consequenceof the tectonic activity (Espinosa and Milla, 2003). This horizon is aresult of the sedimentation of reworked materials: channels aretemporarily flooded by mud stream, which is able to transportendogenous and exogenous remains; this mechanism leads tothe formation of polygenic elements.

6. Conclusion

Calcretes occur on various Quaternary bedrocks such as sand,clay and limestone throughout the tract. The mineralogy of thenon-clay fraction is calcite, quartz and traces of feldspar. Pedo-genic, biotic processes and groundwater may play a role in thedevelopment of calcrete. Palygorskite and calcrete developmentin the eastern Tunisian environment are described as valuable indi-cators of Pleistocene geological paleoenvironments on this plat-form. The first stage seem to be controlled by the formation ofthe nodules taking place through the downward leaching of cal-cium carbonate along primary porosities or secondary porositiesand mainly in shrinkage spaces. Calcretes that developed in associ-ation with soil-forming processes are termed pedogenic forms andusually exhibit beta fabrics with considerable evidences of a bio-logical activity (Wright and Tucker, 1991). Red-bed formation re-quires compulsory oxidising conditions achieved by free drainage(no water logging) (Khadkikar et al., 2000). The red-beds of pedo-genic origin may be classified in the scheme of Mack et al. (1993)as ferric Calcisols. The sand was deposited by flowing water andthe gravel, rolling on the aggrading sandy bed, was preserved ascoating clasts. The grading suggests warning flow conditionsconsistent with deposition from flash floods, typical of wadis(Moumani et al., 2003). This type of calcrete represents the earlystage of calcretisation.

In the stages of calcrete development, the thickness of the cal-careous units is important and the fact that the beds are composed

of a number of distinct subunits suggests that there is a progressiveaggradation, which reveals the action of groundwater fluctuation.This member is a result of the previous material displacement giv-ing rise to the formation of a massive-microlaminar level. Thethickness of this horizon should be controlled by the spatial andtemporal continuity of the subhorizontal root network. (Espinosaand Milla, 2003). This calcareous material develops a network ofplanar centimetric cracks and tubiform pores oriented mainly par-allely to the underlying lamination, and probably generated by theaccommodation of plants roots. This kind of structure reveals adefinitive exundation of sediment, a diagenetic trend leading tohardpan formation and a weathering trend with dissolution of cal-cite and its precipitation in dusty grain-size and partial leach-outfrom the soil profile. The depositional mode is usually by decanta-tion in well drained and oxygenated environment (Vogt, 1984). Thelatest stage characterized by the development of Brecciated hori-zon located in channels near the relief. Channels are recognizableby a trough shape with a gullying base (Elloy and Thomas, 1981)and it was also confirmed by the presence of the oblique bedding.The channel filling deposits are generally not graded polygenicbedding conglomerates (Fig. 3 E). The presence of the brecciatedhorizons indicates a weak width periodic sedimentation (Marriottand Wright, 1993). The laminar horizon, which is a surface deposit,overlies the massive horizon. This level is micritic with the pres-ence of detrital eolian grains. Periodical oscillations favoured theformation of these horizons in wetter periods, while clastic deposi-tion occurred in more arid periods (Alonso-Zarza and Silva, 2002).This case study in Eastern Tunisia highlights the importance ofmany process (pedogenesis, sedimentogenis, climate and tecto-nism) leading to calcrete development.

Acknowledgements

We are grateful to Adel Sghair for his kind assistance in field-work. Thanks are due to Mustapha Ben Haj Ali for providing thenecessary research facilities in National Office of Mines, MohsenJouirou For XRD analysis, Mokhtar Fakhraoui for chemical analysisand Mohsen Hassine for his hospitality. We would like to thank Pr.Zouhair Fekhfekh and Olfa Hachicha for SEM analysis.

References

Alonso-Zarza, A.M., Silva, P.G., 2002. Quaternary laminar horizons with bee nests:evidences of small scale climatic fluctuations, Eastern Canary Islands, Spain.Palaeogeogr. Palaeoclimatol. Palaeoecol 178, 119–135.

Ballais, J.L., Ben Ouezdou, H., 1991. Forms and deposits of the continentalquaternary of the Saharan margin of Eastern Maghreb (tentative synthesis).Journal of African Earth Sciences 12 (1-2), 209–216.

Bédir, M., 1985. Dynamique des bassins sédimentaires du sahel de Mehdia d’aprèsles données de surface: sismostratigraphie, évolutiontectonique etsédimentologie. Mémoire de D.E.A., Fac. Sciences de Tunis, 127p.

Bédir, M., 1988. Mécanismes d’ouverture et évolution tectono-sédimentaire desbassins de transtension néogènes de Tunisie orientale. Rev. Sc. De la terre, vol. 8,pp. 95–108.

Ben Haj Ali, M., Jedoui, Y., Ben Salem, H., Memmi, L., 1985. Carte géologique de laTunisie : 1/500 000. Office National des Mines, Service Géologique.

Ben Jmaa, H., 2008, Le système endoréique de Sidi Al Hani, Chrita et Mechertate:Paléoenvironnement et dynamique récente. Thèse de doctorat, université deTunis.

Bolle, M.P., Adatte, T., Keller, G., Von Salis, K., Burns, S., 1999. The Paleocene–Eocenetransition in the southern Tethys (Tunisia): climatic and environmentalfluctuations. Bull. Soc. Géol. France 170, 661–680.

Botha, G.A., Hughes, J.C., 1992. Pedogenic palygorskite and dolomite in a lateNeogene sedimentary succession, north-western Transvaal, South Africa.Geoderma 53, 139–154.

Bouaziz, S., Barrie, E., Soussi, M., Turki, M.M., Zouari, H., 2002. Tectonic evolution ofthe northern African margin in Tunisia from paleostress data and sedimentaryrecord. Tectonophysics 357, 227–253.

Burollet, P.F., 1956. Contribution à l’étude stratigraphique de la Tunisie centrale.Service des mines de Tunisie, 143p.

Bustillo, M.A., Arribas, M.E., Bustillo, M., 2002. Dolomitization and silicification inlow-energy lacustrine carbonates (Paleogene, Madrid Basin, Spain).Sedimentary Geology 151, 107–126.

456 W. Gallala et al. / Journal of African Earth Sciences 58 (2010) 445–456

Castany, G., 1947. Etude géologique de la bordure occidentale du Sahel. An desmines 2, 134.

Coque, R., 1962. La Tunisie présaharienne : étude géomorphologique. Ed. ArmandColin, Paris, 476p.

Ducloux, J. and Laouina A., 1989. The pendent calcretes in semi-arid climates: anexample located near Taforalt, NW Morocco. CATENA, vol. 16, pp. 237–249(References and further reading may be available for this article. To viewreferences and further reading you must purchase this article.).

Durand, J.H., 1956. Les croûtes calcaires d’Afrique du Nord étudiées à la lumière dela Bio-Rhexistaxie. Serv. Et. Sci. Alger, Pédologie et agrologie 2, 23.

Durand, J.H., Gaucher, G., Lacroix, D., Mathieux, L., Mercier, J.L., Vogt, T., Wilbert, J.,1979. Premières résultats du groupe de travail sur les croûtes calcaires. Bull.Ass. Sén. Et. Quat., 25–29.

Elloy, R., Thomas, G., 1981. Dynamique de la genèse des croûtes calcaires (calcrètes)développées sur des séries rouges pléistocènes en Algérie nord occidentale :Contexte géomorphologique et climatique. Bull. Centres rech. Explor Prod. ElfAquitaine 5, 53–112.

Espinosa, R.J., Milla, J.J., 2003. Calcrete development in Mediterranean colluvialcarbonate systems from SE Spain. Journal of Arid Environments 53, 479–489.

Essefi, E. 2009. Multidisciplinary study of Sidi El Hani Saline Environment: theHistory and the Climatic Variability. Master thesis, Faculty of sciences of Sfax.

Estrela, M.J., Vogt, T., 1989. Etude de croûtes du Quaternaire espanol : importanced’une microflore d’eau douce, comparaison avec le Magreb, t. 308, série II, C.R.Acad. Sci. Paris, pp. 201–206.

Freyet, P., Verrecchia, E., 1989. Les carbonates continentaux du pourtourméditerranéen : microfaciès et milieu de formation. GéologieMéditerranéenne 2–3, 75–83.

Freytet, P., Plaziat, J.C., 1978. Les redistributions carbonatés pédogénétiques (nodules,croûtes, «calcretes»: les deux types principaux d’environnement favorables à leurdéveloppement, t. 286, série D, C.R. Acad. Sci. Paris, pp. 1775–1777.

Gardner, L.R., 1972. Origin of the Mormon Mesa Caliche. Clark County, Nevada. Bull.Geol. Soc. Am. 83, 143–146.

Gauthier-Lafaye, F., Taieb, R., Paquet, H., Chahi, H., Prudencio, I., Sequeira Braga, M.,1993. Composition isotopique de l’oxygène de palygorskites associées à descalcrètes : condition de formation, vol. 316, série II, C.R. Acad. Sci. Paris, pp.817–836.

Isphording, W.C., 1973. Discussion of the occurrence and origin of sedimentarypalygorskitesepiolite deposits. Clays Clay Miner. 21, 391–401.

Jamoussi, F., Bédir, M., Boukadi, N., Kharbachi, S., Zargouni, F., López-Galindo, A.,Paquet, H., 2003. Répartition des minéraux argileux et contrôle tectono-eustatique dans les bassins de la marge tunisienne, vol. 335, C.R. Acad. Sci.Paris, pp. 175–183.

Kamoun, Y., 1981. Etude néotectonique de la région de Monastir-Mahdia (Tunisieorientale). Thèse de Doctorat de 3ème cycle, Paris XI, Orsay, 175p, 12 planches.

Khadkikar, A.S., Chamyal, L.S., Ramesh, R., 2000. The character and genesis ofcalcrete in Late Quaternary alluvial deposits, Gujarat, Western India, and itsbearing on the interpretation of ancient climates. Palaeogeogr. Palaeoclimatol.Palaeoecol. 162, 239–261.

Mack, G.H., James, C.W., Monger, C.H., 1993. Classification of paleosols. Geol. Soc.Am. Bull. 105, 129–136.

Maizels, J., 1990b. Raised channel systems as indicators of palaeohydrologic change:a case study from Oman. Palaeogeogr. Palaeoclimatol. Palaeoecol. 76, 241–277.

Marriott, S.B., Wright, V.E., 1993. Palaeosols as indicators of geomorphic stability intwo Old Red Sandstone alluvial suites, South Wales. J. Geol. Soc. London 150,1101–1120.

Millot, G., 1964. Géologie des argiles. Masson, Paris.Millot, G., Baquet, H., Rudan, A., 1969. Néoformation de l’attapulgite dans les sols à

carapaces calcaires de la Basse Moulouya (Maroc oriental). C. R. Acad. Sci. Paris,268-D, pp. 2771-2774.

Millot, G., Nahon, D., Paquet, H., Ruellan, A., Tardy, Y., 1977. L’épig6nie calcaire desroches silicat6es dans les encroûtements carbonatés en pays sub-aride,Antiatlas, Maroc. Sci. G6ol. 30, 129–152.

Moumani, K., Alexander, J., Bateman, M.D., 2003. Sedimentology of the LateQuaternary Wadi Hasa Marl Formation of Central Jordan: a record of climatevariability. Palaeogeogr. Palaeoclimatol. Palaeoecol 191, 221–242.

Nahon, D.H., Ruellan, A., 1975. Les accumulations de calcaire sur les marneséoceènes de la Falaise de Thiès (Senegal). Mise en évidence de phénomènesd’épigénie. En Colloque Types de croûtes calcaires et leur répartition régionale.7–12. Ed.: T. Vogt. Univ. Louis Pasteur. Strasbourg, 150pp.

Paquet, H., 1983. Stability, instability and significance of attapulgite in the calcretesof mediterranean and tropical areas with marked dry season. Sci. Géol. Mém.72, 131–140.

Patriat, M., Ellouz, N., Deyb, Z., Gauliera, J.M., Ben Kilanib, H., 2003. The Hammamet,Gabès and Chotts basins (Tunisia): a review of the subsidence history.Sediment. Geol. 156, 241–262.

Regaya, K., 2000. L’interface pédogenèse-sédimentogenèse dans les carbonatescontinentaux: contribution à l’étude de lithologie des calcrètes et desformations calcaires associées au Quaternaire de Tunisie. Th. Uni. Tunis II,Fac, Sci, Tunis, 442p.

Regaya, K., 2002. Genèse calcaire associées aux glacis-terrasses de la dorsaletunisienne (piémonts occidentaux de Jebel Bargou). Notes de service géologiquede Tunisie 69, 39–51.

Singer, A., Norrish, K., 1974. Pedogenic palygorskite occurences in Australia. Am.Min. 59, 508–517.

Vanden Heuven, R.C., 1966. The occurrence of sepiolite and attapulgite in thecalcareous zone of a soil near Las Cruces, New-Mexico. In: Clays and ClayMinerals. Proc. 3th Nat. Conf. on Clays and Clay Minerals, Pergamon Press, NewYork, pp. 193–207.

Verrecchia, E., 1994. L’origine biologique et superficielle des croûtes zonaires.Séance spécialisée: carbonates intertropicaux. Bull. Soc. Géol. France, t. 165, n�6,pp. 583–592.

Vogt, T., 1984. Problèmes de genèse des croûtes calcaires quaternaires. Bull. Centresrech. Explor Prod. Elf Aquitaine 8 (1), 209–221.

Wang, Y., Nahon, D., Merino, E., 1994. Dynamic model of the genesis of calcretesreplacing silicate rocks in semi-arid regions. Geochim. Cosmochim. Acta 58,5131–5145.

Watts, N.L., 1980. Quaternary pedogenetic calcrete fro the Kalahari, mineralogy,genesis and diagenesis. Sedimentology 27, 661–687.

Winnock, E., Bea, F., 1979. Structure de la Mer Pélagienne. GéologieMéditerranéenne, 35–40.

Wright, V.P., 1986. The role of fungal biomineralization in the formation of EarlyCarboniferous soil fabrics. Sedimentology 33, 831–838.

Wright, V.P., Tucker, M.E., 1991. Calcretes: an introduction Calcretes. In: Wright,V.P., Tucker, M.E. (Eds.), Int. Assoc. Sedimentol. Rep. Ser. 2, pp. 1–22.