Pedogenic Carbonates As a Proxy for Palaeo-CO2 in the Palaeozoic Atmosphere

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by GEO-LEOe-docs

Pedogenic carbonates as a proxy for palaeo-CO2 in the Palaeozoic atmosphere.

Andres Quast, Jochen Hoefs, Josef Paul

Geowissenschaftliches Zentrum der Universität Göttingen, Goldschmidtstr. 3, D-37077 Göttingen

Abstract

According to a model by Cerling (1991, 1999), the carbon isotope composition of calcretes should depend on the soil type and the CO2-concentration in the atmosphere. We have tested Cerling’s model by investigating 14 Palaeozoic sections with soil profiles. A large number of carbonate types of different genetic origin exist in the localities examined. Comparing the Palaeozoic samples with recent and subrecent calcretes, it can be demonstrated that anhedral, cryptocrystalline (<10 µm) and subhedral microcrystalline (10 - 40 µm) carbonates are clearly of pedogenic origin. Crystals of larger size with a poikilotopic texture are of groundwater or burial diagenetic origin. Macro- and micromorphological features, typical of recent calcretes, occur in several soil profiles, but thin section microscopy reveals a strong diagenetic overprint of most pedogenic carbonates. Time equivalent sections with comparable soil types (protosols, calcisols and vertisols) show large variations in carbon isotope composition. On the other hand, different carbonate generations at one site do not differ much. Therefore Palaeozoic calcretes appear to be unsuitable for a deduction of the Palaeozoic CO2-concentration. Keywords: palaeoatmosphere, carbon dioxide, palaeosols, Palaeozoic, stable isotopes

1 Introduction differences between 13C/12C ratios in atmospheric and plant respired CO2. It describes the relationship Several approaches have been used between both types of CO2 as a func- to estimate the Earth’s atmospheric tion of depth in the soil profile. Be- 13 CO2-concentrations in the geological low a depth of about 50 cm the δ C past. Among them is the GEOCARB- value of soil CO2 is nearly constant model of Berner (1991, 1994) the (Fig. 1). Assuming that pedogenic most prominent. Carbon isotopes in carbonates will retain this isotopic calcretes have been used as another signal, the latter can be used to infer tool to estimate the atmospheric palaeoatmospheric CO2. concentration of CO2 (Cerling 1991, 1999). The Cerling model is based on Cerling (1991) uses a diffusion- ? Published in Palaeogeogr., Palaeocli- production equation to model the matol., Palaeoecol. 2006 contributions of atmospheric and soil Email address: [email protected], respired CO2 in the soil atmosphere. [email protected], [email protected] Davidson (1995) has simplified the (Andres Quast, Jochen Hoefs, Josef equation to estimate the palaeoat- Paul). mospheric CO2 concentration:

http://www.elsevier.com/wps/ﬁnd/journaldescription.cws home/503355/description pedogenic carbonates (Fig. 2). Soil carbonates used for CO2 estimates must have precipitated in the vadose zone in exchange with atmospheric CO2. Carbonates formed in a carbonate hostrock have to be avoided, because of the risk of inherited isotopic signatures.

1.1 Definition and classification of Figure 1. C-isotope ratio of soil-CO2 cal- pedogenic carbonates (calcretes) culated with different respiration rates and paleosols (after Cerling and Quade 1993).

There is no uniform definition of the δ13C − 1.0044 δ13C − 4.4 term calcrete in the literature (e.g. C = S(z) s φ air 13 13 Lamplugh 1902, Netterberg 1967, δ Cair − δ Cs (1) Goudie 1973, Freytet and Plaziat 1982, Machette 1985, Alonso-Zarza where Cair is the concentration of 2003). Netterberg (1967) defined atmospheric CO2, S is the amount calcretes as carbonates precipitated of soil respired CO2 as a function from supersaturated soil- or ground- 13 13 of depth (z) and δ Cs, δ Cair , waters near the soil surface. We use 13 δ Cφ are the isotopic compositions the term groundwater carbonate for of soil-CO2, atmospheric CO2 and all kinds of carbonates precipitated soil respired CO2 respectively. in the phreatic soil zone (Fig. 2). The term calcrete is used exclusively for Equation (1) is suitable as an at- carbonates from the vadose zone. mospheric CO2-palaeobarometer if several assumptions are fulfilled. Iso- Netterberg (1980), Goudie (1983), topic composition of soil CO2 is re- Machette (1985) provided a classifi- flected by soil carbonates, that pre- cation of calcretes based on morpho- cipitate in equilibrium with the soil logical features, which also reflects CO2 and below a depth of 50 cm stages of successive calcrete develop- (Cerling 1991, 1999). Assumptions ment (Fig. 3). Stage 1 to 6 reflect an are required for the concentration increase in maturity. These stages are and isotopic composition of soil almost equivalent to calcretes types. 13 respired CO2 (S(z), δ Cφ) and also for the isotopic composition of the Macro- and micromorphological fab- 13 atmospheric CO2 (δ Cair), which in- rics are a clue of terrestrial carbonate troduces uncertainties in the Cerling origin (Wright and Tucker 1991, Pi- model (Ekart et al. 1999, Royer et al. mentel et al. 1996, Khadkikar et al. 2001). 2000). As it is important to distinguish between pedogenic and ground- For an estimate of palaeoatmospheric water derived carbonates some crite- CO2-concentrations calcretes need to ria are discussed below: be distinguished from other terrestrial carbonates. Palustrine and even An asymmetric vertical distribution lacustrine carbonates, if in nearshore of carbonate within a soil profile is position, can display root traces and one feature of pedogenic carbonates. micromorphological fabrics similar to In most cases a gradual lower and a

2 Figure 2. Diﬀerent kinds of terrestrial carbonate with similar micromorphology (after Freytet and Plaziat 1982)

Figure 3. Idealised scheme of calcrete types and stages of maturity, sources of carbonate in soils. sharp upper boundary of carbonate conditions during carbonate growth precipitation is observed. However in the vadose soil zone (Khadkikar this criterion does not fit in all cases. et al. 2000, Retallack 1990, Wright E.g. calcretes formed under varying et al. 1993). Movement and rotation climatic conditions and in association of concretions due to pedoturbation with a repeated soil development nor- may also lead to granular cracks. mally fail to display an asymmetric profile (Marriott and Wright 1993). If palaeosols are truncated or have ex- perienced repeated soil development, Clotted textures are fabrics of dense the reconstruction of the original (cryptocrystalline) concretions in a depth of carbonate accumulation is coarser matrix of microcrystalline difficult or impossible (Cojan 1999). carbonates. Clotted textures are very common in mature calcretes (hon- Granular cracks are fractures in and eycomb to boulder calcretes, Fig. 3) around carbonate concretions, nor- and thought to be formed by coa- mally filled with micro- to finely crys- lescence of pre-existing nodules in talline carbonates. They are thought a vadose environment (Wright and to reflect alternating wet and dry Tucker 1991).

3 Floating grains are isolated clastic s u r f a c e d i a g e n e s i s grains in a carbonate matrix. They vadose carbonate phreatic carbonate recalcification redolo− are not in contact with other clastic precipitation precipitation R1 mitisation P1− 3 G1 RD grains and are formed by dis- and re- cavity cements aggrading joint− , cavity cements H1 neomorphism Kx, Hx placement of siliciclastic soil compo- S1 i n i c n r

c aggrading nents due to carbonate precipitation e r a

e neomorphism s a i n s Sx i (Goudie 1983). Floating grains are g n

g e

x b h typically found in the vadose zone u u

r dolomitisation m i a l D1 a

(Freytet and Plaziat 1982, Khadkikar t i o et al. 2000). n

Exploding grains are clasts brecciated b u r i a l d i a g e n e s i s and dispersed by carbonate growth. Figure 4. Carbonate types attributed to Often floating grains show such a different diagenetic stages. breakage. Khadkikar et al. (2000) as- sume that water penetrates grains fore are features of the vadose soil along microfractures. Repeated wet- zone (Yaalon and Kalmar 1978, Soil ting and drying in the vadose zone Survey Stuff 1975, Khadkikar et al. supports carbonate precipitation 2000). along microfractures.

Unfortunately, most palaeosols of 1.2 Diagenesis Palaeozoic age are altered by diagenetic processes and many features are lost during geologic history. There- The carbonate types at the different fore, the nomenclature of recent soils localities are described and classified cannot be used without modifica- as representing early, burial and very tions. In this paper the nomenclature late diagenetic stages, where calcretes of Mack et al. (1993) is applied. It and groundwater carbonates are un- utilizes the most prominent macro- derstood as early diagenetic forma- morphological features preserved in tions from the vadose and phreatic a palaeosol. The palaeosols in this environment respectively. Origin of study are classified as protosols, the different carbonate types at each calcisols and vertisols. According locality is discussed below (Fig. 4). to Mack et al. (1993) all of these classes can provide calcretes. Cal- cisols therein are soils, which com- 1.2.1 Carbonate precipitation prise a prominent horizon of carbonate accumulation. Protosols provide Crypto- and microcrystalline carbon- evidence of pedogenesis, including ates Brewer (1964), Knox (1977) and root traces or peds, but they lack Khadkikar et al. (2000) found that any prominent pedogenic horizonta- clotted textures formed by micro- tion. Vertisols may display horizon- crystalline carbonates, that enclose tation but their diagnostic feature small patches of cryptocrystalline is a strong pedoturbation, revealed carbonates, are a common feature by deep tracking desiccation cracks of recent and subrecent calcretes. and pseudo-anticlines, which are The cryptocrystalline carbonates are curved and intersected slip and shear usually anhedral, whereas the sur- planes within the soil profile. They rounding microcrystalline crystals are thought to be formed due to clay often display remarkable rhombohe- shrinking and swelling in seasonally dral shapes (Folk 1971, Wright and wetted and dried soils and there- Peeters 1990). A dull brown or grey

4 occurence of cryptocristalline car- and Mozley 1998). Similar calcites bonates is attributed to an incorpara- are abbreviated as G1. tion of clay minerals during carbonate precipitation in the vadose zone (Wieder and Yaalon 1982). Vadose 1.2.2 Aggrading neomorphism carbonates usually have low Fe- and Mn-concentrations and variable but Aggrading neomorphism describes low Sr-contents from 60 to 220 ppm the alteration of pre-existing carbon- (Khadkikar et al. 2000, Morad et al. ates accompanied by an increase in 1998). For carbonates representing crystal size, but without changes in crypto- and microcrystalline carbon- overall chemistry (Folk 1965). Neo- ats we use abbrevation P1 and P2 morphic carbonates have to be sep- respectively in this study. arated from mesogenetic cements. They often replace small amounts Microcrystalline hem-carbonates of carbonate, whereas mesogenetic Klappa (1983), Wright et al. (1993) cements occur pervasively. and Balin (2000) describe micro- Micro- to coarsely crystalline calcites crystalline carbonates forming rims Neomorphic formation of micro- to around clasts and fingering into sur- coarsely crystalline carbonates is in- rounding carbonates. Although va- dicated by the occurrence of concavo- dose rim cements usually appear convex crystal contacts and by en- on the downsides of clasts, pedo- closures of smaller patches of crypto- genic turnover often leads to rim ce- and microcrystalline crystals (Folk ments that lack orientation (Wright 1965, Bathurst 1975). Corrosive con- et al. 1993). Microcrystalline hem- tacts to clastic grains and jagged carbonates found in the sections are margins to other crystals rule out a termed P3. mesogenetic cement origin for these carbonates (Bathurst 1975, Fucht-¨ Poikilotopic calcites Friedman (1965) bauer 1988). Abbrevation used is S1 describes coarser sized poikilotopic to Sx. cements that fill spaces between sedimentary grains and also enclose sin- gle clasts. Poikilotopic calcites have 1.2.3 Dolomitisation, dedolomitisa- been found from meteoric phreatic tion and redolomitisation but also from deep burial environ- ments (Folk 1974, Scholle 1979, Micro- to coarsely crystalline, rhom- Beckner and Mozley 1998, Adams bohedral dolomites Spötl and Wright and MacKenzie 2001). Low sediment (1992), Garcia et al. (1998) and Wor- compaction within the cemented den and Matray (1998) have demon- parts of rock, indicates an early dia- strated a burial origin of coarser crys- genetic formation of poikilotopic cal- talline, rhombohedral dolomites and cites, which precipitate almost with- ankerites within a sandstone matrix. out affecting stratification (Beckner According to our nomenclature they and Mozley 1998, Garcia et al. 1998). are D1-dolomites. Garcia et al. (1998) and Morad et al. (1998) found low Fe-concentrations Crypto- to microcrystalline, anhedral in early diagenetic poikilotopic cal- calcites Clark (1980), Judersleben cites. Poikilotopic calcites with a uni- and Voigt (1993), Al-Hashimi and form appearence under cathodolumi- Hemingway (1973) found patches nescence, are formed in the phreatic of crypto- to microcrystalline cal- rather than the vadose zone (Beckner cites that imitate coarser rhombo-

5 hedral crystals and therefore in- 20 to 30 million years. One or more dicate a recalcification of former stratigraphic sections were measured dolomites. Rhombohedral hydrox- and described from each time inter- ide rims may indicate a pre-existing val in order to provide data fromEd- dolomite too (Richter 1974, Richter inborough comparable paleosol mor- 1985). A late near-surface recalcifica- phologies across the entire time scale tion is assumed for such carbonates studied. Sections were sampled at (Al-Hashimi and Hemingway 1973, depths comprising carbonate con- Morad et al. 1995). Low magnesium cretions or horizons. Samples were calcites (LMC) with low Fe-content cut into 2 halves and one piece was are in agreement with late near- polished. Powders for stable isotope surface recalcification. Calcites of measurements (10 to 20 mg) were similar origin are abbreviated with obtained from the polished half by R1. micro-drilling. A binocular microscope has been used to drill patches Crypto- to microcrystalline, an- of 2 to 5 mm with optically uniform hedral and microcrystalline, subhe- carbonates, which was verified by dral dolomites that imitate coarser thin sections taken from the other rhombs are thought to be redolomites half of sample (Table 1). (Clark 1980). Some larger crystals floating within the crypto- to mi- Carbon and oxygen isotope ratios crocrystalline matrix can testify a have been measured with a Finnegan former presence of coarser euhe- MAT 251 by the method of Mc- dral dolomites. Abbreviation for this Crea (1950). Extractions were run 24 dolomite type is RD. hours at 25◦ C. Carbon and oxygen isotope ratios are given in the standard permil notation with respect to 1.2.4 Carbonate cements in frac- the PDB standard. Inhouse working tures and caves standard Solnhofen 1 has been used routinely. Analytical uncertainty is Micro- to coarsely crystalline, sub- 0,1% for δ13C and 0,2% for δ18O. hedral carbonates Cements in root The phoshoric acid liberation of CO2 traces and early solution cavities are produces differences in oxygen iso- often thought to be early diagenetic tope composition from calcite and in origin because of the preserva- dolomite that have to be considered tion of the cavity structures (Retal- (Rosenbaum and Sheppard 1986, lack 1990). Other micro- to coarsely Sharma et al. 2002, Hoefs 2003). Be- crystalline, subhedral calcites can be cause there is no consensus about the found within post burial fractures correct fractionation factor and the and caves. Abbreviation for similar samples investigated are often mix- carbonates is H1 to Hx and Kx re- tures of both carbonates, we give raw spectively. data without correction.

A Zeiss Axiolab microscope with a 2 Methods Citl cathode CL8200MK3A is used for cathodoluminescence observations. Element analysis is performed 14 localities with palaeosols from with a Jeol JXA 8900 RL electron Lower Devonian to Lower Permian microprobe. Measured concentra- age were selected (Fig. 5). The lo- tions were automatically corrected calities are assigned to three time with the software CITAF (Arm- intervals. Each interval comprises of strong 1995). Qualitative and quan-

6 Figure 5. Stratigraphical and geographical position of the examined localities titative carbonate analyses were Site P1−3 G1 S1 Sx D1 R1 RD Hx, Kx Llansteffan O X X X X X X performed by staining thin sections Lydney O X X X Ffairfach O with Alizarin-S and potassium fer- Carnousite O Campsie X X X X rocyanid, X-ray diﬀractometry and Aberdour O coloumetric carbonate content de- Pease Bay O X termination (Evamy and Shearman Rock Hall O X Vatterode X X X 1962, Dickson 1966, Muller¨ and Gast- Rothenburg O X X X X X X Kyffhäuser O ner 1971). A Philips PW 1800 is used Hainichen X X X Well De Lutte 6 O X X for X-ray diﬀractometry. Calcite and Lodève O X X X: carbonate type was found at this site dolomite contents are determined O: former occurrence of carbonate type is indicated semiquantitatively by comparison of by sedimentological evidence the (104)-calcite and d(104)-dolomite Figure 6. Carbonates types found or in- peaks using laboratory standards ferred from sedimentological evidence at (Faupel and Thomson 1989). Ca/Mg- each localitiy. Abbreviations for carbon- ratio of dolomite is calculated by us- ate types are given in the text. Hx- and ing the d(104)-dolomite peak (Lums- Kx-carbonate types are attributed to den 1979). early and very late diagenetic stages. marises the carbonate types observed or inferred by sedimentological evidence at each locality. 3 Descriptions and interpreta- tions of sites and soils 3.1 Lower Devonian sites

The investigated localities were selected from recent literature (Ta- The Welsh sites Llansteffan, Lydney ble 1). Detailed field work has re- Harbour and Ffairfach contain pro- vealed, however, that some sites tosols, vertisols and calcisols within provide no suitable calcrete or even fluvial sediments of a coastal plain palaeosols (Quast 2003). The car- (Allen 1964, Friend and Williams bonate types found at the localities 1978). Various soils are developed as are described from the locations were indicated by pseudo-anticlines, root they are best examined. Fig. 6 sum- traces and an upward loss of strat-

7 Table 1 Localities of paleosols and their classification oßler¨ (1995) orner¨ et al. (2001) K Trewin (1987), Balin (2000) Gebhardt (1988), Schneider and Gebhardt (1993) Gebhardt (1988), Schneider and Gebhardt (1993) Paul (2004) Schneider and R Pagnier and v. Tongeren (1996) Wright et al. (1993) References Allen (1964), Friend and Williams (1978) Allen (1964), Friend and Williams (1978) Quast (2003) Friend and Williams (1978) Friend and WilliamsSaigal (1978), and Walton (1988) Friend and WilliamsTrewin (1978) and , Kneller (1987) TS, XRD TS, XRD TS, CL, EMS, XRD TS, EMS, XRD TS Methods TS, CL, EMS, XRD XRD TS, CL, EMS, XRD XRD — NC, HP — NC NC, HC, HP, BC NC NC, HC NC NC NC — NC Calcrete stages* NC, HC, HP NC NC, HC NC, HP NC NC NC NC, HP, BC NC, HP NC NC, HC NC 1 Calcisol 2 Protosols 3 Calcisols, 1 Protosol 1 Protosol 2 Calcisols, 3 Protosols 1 Calcisol 1 Protosol 5 Protosols 2 Protosols 2 Protosols, 2 Vertisols 1 Protosol 1 Vertisol 1 Calcisol, 1 Vertisol 1 Protosol, 1 Vertisol 1 Protosol ? 1 Protosol 2 Calcisols 1 Protosol, 1 Calcisol 1 Protosol 1 Caclisol 1 Protosol ? Soil Alluvial-fan and -plain Alluvial-fan and -plain Alluvial-fan and -plain Alluvial-fan and -plain Alluvial-plain Alluvial-plain Alluvial-fan and -plain Coastal-plain Coastal-plain Coastal-plain Alluvial-fan Alluvial-plain Alluvial-fan and -plain Alluvial-fan and -plain Alluvial-fan and -plain Environment artensdorf Fm. ¨ Untere Mansfeld Subgroup Untere Mansfeld Subgroup Untere Mansfeld Subgroup H Limburg Fm. Viala Fm. UORS Red Marl Fm. Raglan Marl Group Red Marl Fm. Garvock Group Strath Rory Group Garvock Group Crovie Group Kinnesswood Fm. Lithostratigraphy R 44 6039/ H 57 1835 R 44 8294/ H 57 2278 R 44 3561/ H 56 9844 — — — NO 768 647 SN 350 099 SO 512 741 SN 191 610 NO 124 328 NO 570 345 NJ 658 884 NT 792 721 National Grid 2 2 2 4 2 3 2 2 3 2 auser ¨ Vatterode Rothenburg 1 Kyffh Hainichen Core De Lutte 6 Lodève 1 Rock Hall 1 Llansteffan 1 Lydney Habour 1 FfairfachCampsie 1 Carnoustie 1 Aberdour Beach 1 Pease Bay 3 Locality/ Section *Abbreviations given for calcrete- stages and EMS: analytical electron methods: NC: microprobe nodular analysis calcrete - - HC: XRD: honeycomb X-ray calcrete - diffractometry HP: hardpan - BC: boulder calcrete, TS: thinsections - CL: cathodoluminescence

8 pseudo− anticlines

small nodular calcretes nodular calcretes and root traces

honeycomb calcretes nodular calcretes carbonates derived from the under− lying calcretes

calcretes gravels root traces

A B

Figure 7. Well developed paleosols A) protosol in section Llansteffan 2 with light brown horizontal aligned calcrete gravels at its base, a horizon comprising carbonates derived by dissolution of the latter and some small in situ calcretes near the top of soil. B) vertisol at the section Llansteffan 3 comprises root traces, pseudo-anticlines and variuos calcrete stages at different soil levels. Scale is 1 m at each section. ification. Pedogenic, groundwater, file, a pedogenic origin is suggested. reworked and diagenetic carbonates In contrast to section 2 the section were found (Fig. 7). Section Llanst- Llansteffan 3 provides a vertisol with effan 2 for example displays carbon- carbonates exclusively of pedogenic ates from different origin within a origin. Carbonate precipitation is re- protosol. At the base of the proto- stricted to the vadose zone, which is sol, horizontal and lenticular con- indicated by pseudo-anticlines and a cretions occur within well bedded loss of stratification. sediments (Fig. 7A). Their occurrence and their vertical position in Although sedimentological evidence the profile excludes vadose formation for pedogenesis is given for all local- within the soil. On the other hand, ities, no P1 to P3 carbonates have they display granular cracks, float- been found at these sites. At Llanst- ing grains and sharp margins, often effan micro- to finely crystalline Sx- observed in pedogenic carbonates calcites occur near stylolites and from other sites. Therefore, rework- reveal a length growth normal to ing, abrasion and fluvial transport is stylolithes solution planes. Contem- suggested for these carbonates, sim- poreous formation of Sx-calcites and ilar to the calcrete gravel Marriott stylolites is therefore postulated. Sty- and Wright (1996) have described lolites indicate a mesogenetic stage at this locality. Above the calcrete with moderate burial (Clark 1980, gravels isometric concretions occur. Bathurst 1983, Langbein et al. 1982). Upwards, the number and size of con- A second type of micro- to coarsely cretions decrease rapidly. They were crystalline Sx-calcites enclose patches formed most likely by dissolution of pre-existing carbonates. Carbon- of the underlying calcrete gravels. ate staining delineates the coexis- Above a zone that lacks any concre- tence of calcite and dolomite within tions, smaller concretions appear in the patches. Therefore a rather com- the upper part of protosol together plex carbonate generation is consid- with root traces. They are sharply ered for the micro- to coarsely crys- bounded to the sediment. Due to talline Sx-carbonate. It is a LMC with their vertical position in the soil pro- minor Fe-content (0.2%, Table 2).

9 Table 2 Element microprobe analysis results and calculated carbonate compositions

Locality Carb. n Ca CaCO3 Mg MgCO3 Fe FeCO3 Mn Sr type (wt%) σ (mol%) (wt%) σ (mol%) (ppm) σ (mol%) (ppm) σ (ppm) σ Campsie P1 10 38.7 0.8 97.3 0.5 0.2 1.9 1185 659 0.2 3328 3288 313 222 P2 13 40.0 0.7 98.8 0.2 0.1 0.9 441 387 0.1 810 1103 143 227 P3 10 39.7 0.8 97.3 0.3 0.1 1.2 717 898 0.1 7793 6256 344 269 G1 6 39.5 0.4 98.3 0.2 0.1 0.7 570 258 0.1 4637 2091 1025 1288 S1 5 40.4 0.2 99.2 0.1 0.0 0.5 n.n. – 0.0 1316 893 198 86 Hx 4 39.8 0.8 97.7 0.2 0.1 0.9 2024 3163 0.4 6095 2850 197 107 Kx 9 39.5 0.9 96.8 0.3 0.2 1.3 2675 3169 0.5 7685 2786 371 270 Llansteﬀan Sx 3 38.0 0.6 96.9 0.6 0.2 2.5 1114 82 0.2 1685 338 539 72 Hx 1 39.5 – 97.7 0.2 – 0.8 n.n. – 0.0 8522 – 110 – Kx 4 39.5 0.3 98.0 0.2 0.1 0.7 775 752 0.1 5647 3196 1557 1118 Rothenburg R1 6 39.4 0.8 98.4 0.3 0.1 1.3 1428 1402 0.3 237 246 172 129 Hx 3 40.2 0.3 99.3 0.2 0.1 0.6 176 249 0.0 31 38 175 82 Vatterode P1 9 21.8 0.3 50.6 12.5 0.4 47.8 2399 594 0.4 4960 3156 1847 848 P2 4 21.8 0.2 51.0 12.5 0.2 48.2 1296 358 0.3 2419 2424 1572 794 Hx 2 40.7 0.1 98.4 0.3 0.1 1.0 n.n. – 0.0 3421 488 64 39

Although such low Fe-content is un- matrix. Vertical arrangement and an usual for carbonates formed during increase in size towards the top indi- burial diagenesis, Saigal and Bjør- cate a pedogenic origin of concretions lykke (1987) have found similar cal- at Carnoustie and Aberdour Beach. cites from a shallow burial diagenesis. At Lydney Harbour fine to medium At section Campsie 3 pedogenic over- crystalline dolomite rhombs have re- print is easier recognizable (Fig. 8). placed pedogenic and non-pedogenic At its base fine grained sediments be- carbonates. Internal fabrics were de- come increasingly destratified, root stroyed in almost all calcretes, but traces are found at higher levels. The ghost structures reveal that car- protosol lacks any carbonate concre- bonate originated from pedogenesis. tions, but is completely cemented Staining indicates low Fe-dolomites. by a poikilotopic calcite. Well strat- Similarly low Fe-dolomites from Tri- ified layers follow above the soil and assic sandstones have been formed have cut deep channels in it. A few during burial diagenesis (Morad et al. root traces indicate a weak soil de- 1995). velopment. Carbonate concretions coalesce laterally into a hardpan. Granular cracks, clotted textures, Campsie, Carnoustie and Aberdour floating and exploded grains reveal Beach (Midland Valley, Orcadian the pedogenic origin of carbonates. Basin, Scotland) contain calcisols and protosols from an alluvial-fan P1, P2 and P3 carbonates are found and -plain facies. Most sections from at all Campsie sections. Thin sec- these sites lack root traces or peds, tions exhibit small patches of cryp- but macro- and micromorphological tocrystalline, anhedral carbonates appearance of carbonates suggests a (P1), which are enclosed by micro- pedogenic origin. At Campsie 1 and crystalline, but subhedral carbonates 2 thick limestone horizons appear (P2). Their primary rhombohedral as coalesced carbonate concretions shape is detected by cathodolumi- comprising granular cracks and clot- nescence (Fig. 9A). A pale orange ted textures. Some floating and ex- (Campsie) occurrence of P1 carbon- ploded grains suggests a strong dis- ates is attributed to the incorpora- and replacement of a former clastic tion of clay-minerals during vadose

10 Figure 8. Overview of sections Campsie 1 and Campsie 3

P2 Fsp

P1 Qz P1 P2 G1

A B C

Figure 9. Carbonate types from early diagenetic origin. A) and B) are cathodoluminescence images. A) Patches of cryptocrystalline calcites (P1) are enclosed by microcrystalline calcites (P2) in the calcretes of the Campsie site. The latter reveal rhombic shapes by cathodoluminescence. B) P1 and P2 carbonates similar to those of the Campsie site consist of dolomite in the mature calcretes at Vatterode. C) Sedimentary grains are enclosed by a poikilotopic calcite (G1) at Campsie section 3

11 carbonate precipitation. P1, P2 and soil formation and carbonate mor- P3 carbonates from Campsie are phology suggest a groundwater ori- low-Mg-calcites (LMC) with also low gin for these concretions and layers. Fe-concentrations. Sr-contentrations Only at the sections Pease Bay 4 and range from 200 to 400 ppm, com- Rock Hall 2 some concretions are parable to recent calcitic calcretes. correlated with weak soil develop- Early diagenetic G1 carbonates have ment indicated by desiccation cracks completely cemented the lower part (Pease Bay 4), blocky aggregates of section Campsie 3 (Fig. 9C). and root traces. These concretions Cathodoluminescence reveals uni- differ from groundwater carbonates form bright orange crystals. Poik- by their sharp margins and a strong ilotopic calcites are generally LMC, displacement of clastic grains. with less than 1% Mg and less than 0.2% Fe. They generally contain much higher Sr-concentrations than 3.3 Permo-Carboniferous sites the vadose P1 to P3 carbonates. The non-pedogenic S1, Hx, Kx carbonates from Campsie are LMC with Profiles which are increasingly de- Sr-concentrations that do not differ stratificated upwards and partly ac- significantly from the pedogenic car- companied by root traces and des- bonates. Iron is below the detection iccation cracks gives evidence for limit within S1 carbonates, whereas pedogenic development at the sec- Hx and Kx carbonates reveal Fe- tions from Rothenburg, Kyffhäuser concentrations of about 0.4% Fe, and (Saale Basin, Germany), De Lutte 6 therefore higher than those of pedo- (Ems-Graben, Netherlands), Lodève genic carbonates. (France). Micromorphology supports a calcrete origin for most carbonate concretions at these sites and 3.2 Upper Devonian to Lower Car- also at Hainichen (Erzgebirge Basin, boniferous Sites Germany). At Rothenburg besides some calcretes, layered concretions occur, interpreteted as groundwater Pease Bay and Rock Hall (Midland carbonates. Valley, Scotland) both exhibit thick carbonate horizons in alluvial-fan Vatterode (Saale Basin, Germany) and -plain deposits of an intramon- exhibit three calcisols and a pro- tane basin. Most carbonate concre- tosol. Calcretes within the soils re- tions occur within trough cross bed- veal very well preserved microfab- ded deposits that lack any evidence rics (Chap. 1.2). Because of its good of pedogenic overprint (Fig. 10A). preservation a more detailed descrip- The concretions are often arranged tion is given: parallel to bedding planes and reveal gradual transitions into the sed- Pedogenic overprint is testified at iments. They are formed by poikilo- Vatterode by an upward destratifica- topic calcites and lack internal fab- tion and minor root traces. Carbon- rics like granular cracks or clotted ate contents of the calcisols typically textures. Eventually, concretions co- increase towards the top. Calsisol 2, alesced into layers, without affecting for example, exhibits a siltstone with significantly the former stratification randomly distributed calcite concre- (Fig. 10B). Only minor displacement tions at its base (Fig. 11). Stratifi- of clastic grains is seen in the carbon- cation is generally intact but is de- ate concretions and layers. Missing stroyed near the concretions. The

12 A B

Figure 10. A) Carbonate concretions at Pease Bay are correlated with coarse grained, cross bedded sandstones that lack any evidence of pedogenic overprint. B) Carbonate has cemented several horizons of sandstones without disturbance of cross stratification at the Rock Hall site. Note the concretions are similar to those of Pease Bay, which are fused together at the base of the cemented layer and also carbonate sheets imitating bedding planes at the top of layer. Lens cap in A) is 5 cm in diameter, scale given in B) is 1 m. amount of concretions increases up- The large number of calcretes ac- ward and siltstone passes into non- companied by moderate pedogenic continuous carbonate layers. Carbon- overprint of surrounding sediments ate mineralogy changes from calcite indicates soil formation in an arid cli- to dolomite. Carbonate concretions mate. This assumption is supported coalesce into an overlying 40 cm thick by an early dolomite precipitation impermeable dolostone horizon. The and a silcrete horizon, within the soil is upward terminated by a sec- fourth soil profile (Fig. 11). The prin- ond dolostone horizon, where coa- cipal carbonate precipitation should lesced concretions form well rounded be directly below the soil surface at boulders, separated by small joints arid climates (McFadden and Tinsley of clastic material. 1985). Therefore the tops of boulder calcrete and hardpans, respectively, represent a palaeosol surface, which The succession very well compares is important for the interpretation of with calcrete stages found in modern stable isotope data (Chap. 4). Car- calcisols (Machette 1985). It encloses bonate concretions are found directly nodular, honeycomb and hardpan above the hardpan of the third cal- calcretes. The uppermost horizon cisol, thought to be derived by hard- corresponds to a boulder calcrete pan dissolution, which occurs when formed at the beginning of brec- the soil is covered by further deposits. ciation and dissolution of a former In addition to the soil carbonates hardpan (Goudie 1983; Wright and the section includes lacustrine lime- Tucker 1991). Carbonate concretions stones. in the calcisol reveal sharp margins to the sediment, granular cracks and clotted textures. Strong re- and Cathodoluminescence reveals P1 displacement of clastic grains, due and P2 carbonates at Vatterode, to carbonate precipitation, is sug- very similar to those of Campsie gested by rare floating and exploded (Fig. 9A/B). The carbonates are grains. Boulders of the highest hori- LMC and stoichiometric dolomites. zon partly exhibit laminae at their Both, calcites and dolomites are at- upside margins, which is often found tributed to formation in the vadose in mature calcrete profiles (Goudie environment. Evaporation accom- 1983, Machette 1985) panied with strong precipitation

13 Figure 11. Overview of the Vatterode section (legend see Figure 8). of LMC leads to an enrichment of crypto- to microcrystalline dolomites Mg in soil water which may trig- (RD) imitate coarser rhombs, which ger dolomite precipitation (Watts are coated by hydroxides in many 1980). Therefore dolomite is found instances. Some larger subhedral predominantly in mature calcrete dolomites are floating within the stages at Vatterode (Fig. 8). How- crypto- to microcrystalline matrix, ever, direct dolomite precipitation which underlines the former pres- from pore waters is under debate ence of coarser euhedral dolomites. A and formation from a former calcite near-surface fault bound redolomiti- must be taken into account (Arakel sation is assumed. and McConchie 1982). Low Fe- and Mn-concentrations of P1 and P2 carbonates are attributed to car- 4 Stable isotope compositions bonate formation in a vadose environment (Morad et al. 1998). High Sr-concentrations of about 1500 and The localities, we examined, con- 1800 ppm found in P1- and P2- tain protosols, calcisols and vertisols. dolomites, are explained by equi- All sites were selected from basins librium with evaporitic pore waters near the palaeo-equator. According (Morad et al. 1998). Hx carbonate to the Cerling model, such palaeosols from Vatterode has no detectable Sr of nearly the same age, should give and also lacks Fe. comparable carbon isotope ratios from pedogenic carbonates, which Micro- to coarsely crystalline, rhom- is, however, not the case (Fig. 12). bohedral dolomites (D1) occur at the At Lower Devonian and Devono- Lodève sections. Similar to those of Carboniferous sites δ13C-values from Lydney Harbour they replace car- calcretes differ by 13% and 7.5% re- bonates in a facies independent man- spectively (Fig. 13). Those of the ner and thought to be formed by deep Permo-Carboniferous sites reveal a burial diagenesis. They are enriched difference of 6.7%. Because of the in Ca (pers. communication Frank rapid evolution of vascular plants in Körner). At the section Rotenburg 1 the Lower Devonian, localities in this

14 age range perhaps must be divided genic carbonates are enriched in 18O into shorter time intervals (Berner compared to groundwater carbonates 1997, 1998). If sections of Lochko- and cavity or fracture fills. This rela- vian age are separated from those tionship suggests the preservation of of Pragian and Emsian age, isotope carbon isotope compositions of pedo- variations in each interval become genic carbonates at the Campsie site. smaller, but still have differences of about 5%. Variations in this range The Vatterode site is the only site, 13 lower the accuracy of CO2 estima- that exhibits an increase of δ C- tions considerably. values upwards in a soil profile. Iso- tope data from Vatterode, there- Especially for the Lower Devonian fore, obviously match the Cerling sites, each locality has a distinct car- assumptions, but the upward change bon isotope composition that does of carbon isotope is associated with not match any other locality. At all a change from calcite to dolomite sites, non-pedogenic carbonates re- (Fig. 11). Due to its higher δ18O- veal a carbon isotope composition values a pedogenic formation of close to or identical to soil derived dolomite seems likely (Land 1980). carbonates, although different values Nevertheless it is not clear if 13C- might be expected due to the differ- values of dolomites represent 13C- ent carbonate origin (Fig. 12). values of soil CO2 at higher soil levels. A dolomite specific 13C-enrichment With some exceptions, O-isotope val- cannot be excluded. In coincidence ues for pedogenic and non-pedogenic with most other sites, Vatterode re- carbonates are more or less the veals no significant difference in δ13C- same at each site. Most sites reveal values, when comparing the pedo- δ18O-values for pedogenic and non- genic carbonates with non-pedogenic pedogenic carbonates that match carbonates. meteoric to shallow burial conditions, whereas high δ18O-values from Lyd- In summary, diagenesis appears to be ney Harbour and Rothenburg sites the most important process in affect- may be derived from dolomitisation ing the isotope composition of pedo- (Chap. 1.2). At Ffairfach primary pe- genic and non-pedogenic carbonate. dogenic carbonates have δ18O-values A detailed discussion of the different typical for burial diagenesis, whereas diagenetic stages appears to be nec- fracture fillings reveal compositions essary. that favor a meteoric-water origin.

By comparison with the microstruc- 5 Conclusions tures from modern calcretes, pedogenic carbonates of the Campsie and Vatterode sites appear to be nearly Palaeosols from 14 localities near unaltered by diagenesis. If original the palaeo-equator can be classified carbon isotope composition is pre- on the basis of sedimentological fea- served with in these calcretes, a tures. The carbonates can be char- clear distinction between carbonates acterized by their macro- and micro- which formed under different condi- morphology. Most carbonates of an tions maybe delineated, but no sig- early vadose origin appear in asym- nificant δ13C differences are observed metric profiles, which agrees with the at Campsie. In contrast, δ18O-values observations of Goudie (1983) and of the different carbonate types show Wright and Tucker (1991). In con- clear differences at this site. Pedo- trast groundwater carbonates pro-

15 Figure 12. Carbon and oxygen isotope compositons from all localities

Figure 13. Carbon isotope compositions of calcretes through the Paleozoic. Results of this study compared with compilation given in Ekart et al. (1999).

16 vide a rather symmetric occurence work was funded by the Deutsche in vertical succession. Micromorpho- Forschungsgemeinschaft (Ho 375/20- logical evidence for vadose carbonate 2), which is greatful acknowledged. growth is given by simultaneous re- and displacement and cracking of clastic grains. Calcrete nodules display sharp margins with granular cracks, whereas phreatic conditions References cause a lesser sediment displacement and more continuos margins of con- Adams, A. E., MacKenzie, W. S., cretions. Primary non-pedogenic car- 2001. A Colour Atlas of Carbonate bonates are of groundwater, palus- Sediments and Rocks Under the trine or lacustrine origin. Microscope. Manson, London. Al-Hashimi, W. S., Hemingway, J. E., Although a rather conservative be- 1973. Recent dedolomitization and haviour of fossil calcretes has been the origin of the rusty crusts predicted by many workers, thin of Northumberland. J. sediment. sections, cathodoluminescence and Petrol. 43, 81–92. electron microprobe analyses indi- Allen, J. R. L., 1964. Studies in fluvi- cate a complex diagenetic overprint atile sedimentation: six cyclothems for most sites. Palaeosols from three from the Lower Old Red Sand- time-slices, each of which comprise 20 stone, Anglo-Welsh Basin. Sedi- to 30 million years, show large ranges mentology 3, 163–198. in carbon isotope composition. Large Alonso-Zarza, A. M., 2003. Palaeoen- carbon isotope variations in time vironmental significance of palus- equivalent calcretes from different trine carbonates and calcretes. sites do not match the assumptions Earth Sci. Rev. 60, 261–298. of the Cerling model. Therefore the Arakel, A. V., McConchie, D., 1982. palaeosols investigated in this study Classification and genesis of cal- appear unsuitable for estimanting crete and gypsite lithofacies in pa- the CO2 concentration in the Palaeo- leodrainage systems of inland Aus- zoic atmosphere. tralia and their relationship to carnotite mineralization. J. sediment. Petrol. 52, 1149–1170. Armstrong, J. T., 1995. CITAF: A Acknowledgements package of correction programs for the quantitative electron microbeam x-ray analysis of thick pol- We thank C. Wesling (Bochum), for ished materials, thin films, and par- constructive suggestions, technical ticles. Microbeam Anal. 4, 177– support and who also contributed 200. some of the figures, V. Bullwinkel and Balin, D. F., 2000. Calcrete mor- C. Huhne¨ (Göttingen) for stimulat- phology and karst development in ing discussions at each phase of this the Upper Old Red Sandstone at work. Core samples of the De Lutte Milton Ness, Scottland. In: Friend, 6 well were kindly provided from P. F., Williams, B. P. J. (Eds.), the Netherlands Institute of Applied New Perspectives on the Old Red Geosience TNO -National Geological Sandstone. Vol. 180 of Spec. Publ. Survey. We especially thank A.M. Geol. Soc. London, pp. 485–501. Alonso-Zarza (Madrid) and N. Tabor Bathurst, R. G. C., 1975. Carbon- (Davis) for their careful and con- ate sediments and their diagenesis. structive reviews of this paper. This Vol. 12 of Devel. Sediment.

17 Bathurst, R. G. C., 1983. Neomor- 217–231. phic spar versus cement in some Clark, D. N., 1980. The diagenesis Jurassic grainestones: significance of Zechstein carbonate sediments. for evaluation of porosity evolution Contr. Sediment. 9, 167–203. and compaction. J. Geol. Soc. Lon- Cojan, I., 1999. Carbonate-rich don 140, 229–237. palaeosols in the Late Cretaceous- Beckner, J. R., Mozley, P. S., 1998. Early Palaeogene series of the Origin and spatial distrubution Provence Basin (France). In: of early vadose and phreatic cal- Thiry, M., Simon-Coin¸con, R. cite cements in the Zia Formation, (Eds.), Palaeoweathering, Palaeo- Albuquerque basin, New Mexico, surfaces and Related Continental USA. In: Morad, S. (Ed.), Carbon- Deposits. Vol. 27 of Spec. Publ. ate Cementation in Sandstones. Int. Assoc. Sediment., pp. 43–60. Vol. 26 of Spec. Publ. Int. Assoc. Davidson, G. R., 1995. The sta- Sediment., Oxford, pp. 27–51. ble isotopic composition and mea- Berner, R. A., 1991. A model for at- surement of carbon in soil CO2. mospheric CO2 over Phanerozoic Geochim. Cosmochim. Acta 59, time. Amer. J. Sci. 291, 339–376. 2485–2490. Berner, R. A., 1994. Geocarb II: a Dickson, J. A. D., 1966. Carbon- revised model of atmospheric CO2 ate identification and genesis as over Phanerozoic time. Amer. J. revealed by staining. J. sediment. Sci. 294, 56–91. Geol. 36, 491–505. Berner, R. A., 1997. The rise of plants Ekart, D. D., Cerling, T. E., Mon- and their effect on weathering and tañez, I. P., Tabor, N. J., 1999. atmospheric CO2. Sci. 276, 544– A 400 million year carbon iso- 546. tope record of pedogenic carbon- Berner, R. A., 1998. The Carbon cy- ate: implications for paleoatmo- cle and CO2 over Phanerozoic time: spheric carbon dioxide. Amer. J. the role of land plants. Phil. Trans. Sci. 299, 805–827. Roy. Soc. London B353, 75–82. Evamy, B. D., Shearman, D. J., 1962. Brewer, R., 1964. Fabric and mineral The application of chemical stain- analysis of soils. Wiley and Sons, ing techniques to the study of di- New York. agenesis in limestones. Proc. Geol. Cerling, T. E., 1991. Carbon dioxide Soc. London 1599, 102. in the atmosphere: evidence from Faupel, W., Thomson, E., 1989. Der Cenozoic and Mesozoic paleosols. Korallenoolith von Salzhemmen- Amer. J. Sci. 291, 377–400. dorf: Sedimentation, Diagenese Cerling, T. E., 1999. Stable carbon und epigenetische Dolomitisierung, isotopes in palaeosol carbonates. mit geologischer Kartierung. Mas- In: Thiry, M., Simon-Coin¸con, R. ter’s thesis, Universität Göttingen, (Eds.), Palaeoweathering, Palaeo- Göttingen. surfaces and Related Continental Folk, R. L., 1965. Some aspects of Deposits. Vol. 27 of Spec. Publ. Int. recrystallisation in ancient lime- Assoc. Sediment., pp. 43–60. stones. Soc.econ. Paleont. Miner., Cerling, T. E., Quade, J., 1993. Sta- Spec. Publ. 13, 14–48. ble Carbon and Oxygen Isotopes in Folk, R. L., 1971. Caliche nodule com- Soil Carbonates. In: Swart, P. K., posed of calcite rhombs. John Hop- Lohmann, K. C., McKenzie, J., kins Univ. Stud. Geol. 19, 167–168. Savin, S. (Eds.), Climate Change Folk, R. L., 1974. The natural his- in Continental Isotopic Records. tory of crystalline calcium carbon- Vol. 78 of Geophys. Monogr. pp. ate: effect of magnesium content

18 and salinity. J. sediment. Petrol. Academic Press, London, pp. 93– 44, 40–53. 132. Freytet, P., Plaziat, J.-C., 1982. Con- Hoefs, J., 2003. Stable Isotope Geo- tinental carbonate sedimentation chemistry. Springer, Berlin. and pedogenesis - Late Creata- Judersleben, G., Voigt, E., 1993. ceous and Early Tertiary of South- Diagenese und Speicherentwick- ern France. Contr. Sediment. 12, 1– lung des Staßfurtkarbonats am 231. Zechsteinbeckenrand Sudbranden-¨ Friedman, G. M., 1965. Terminology burgs. Geol. Jb. A131, 271–303. of crystallisation textures and fab- Khadkikar, A. S., Chamyal, L. S., rics in sedimentary rocks. J. sedi- Ramesh, R., 2000. The charac- ment. Petrol. 35, 643–655. ter and genesis of calcrete in Friend, P. F., Williams, B. P. J. Late Quarternary alluvial de- (Eds.), 1978. International Sym- posits, Gujarat, western India, and posium on the Devonian System its bearing on the interpretation (P.A.D.S. 78): A field giude to se- of ancient climatic. Palaeogeogr., lected outcrop areas of the Devo- Palaeoclim., Palaeoecol. 162, 239– nian of Scottland, the Welsh Bor- 273. derland and South Wales. Paleont. Klappa, C. F., 1983. A process- Assoc., London. response model for the formation Fuchtbauer,¨ H., 1988. Sedimente of pedogenic calcretes. In: Wilson, und Sedimentgesteine. Sediment- R. C. L. (Ed.), Residual Deposits: Petrologie Teil II. Schweizerbart, Surface Related Weathering Pro- Stuttgart. cesses and Materials. Vol. 11 of Garcia, A. J. V., Morad, S., De Ros, Spec. Publ. Geol. Soc. London, pp. L. F., Al-Aasm, I. S., 1998. Palaeo- 211–220. geographical, palaeoclimatic and Knox, G. J., 1977. Calcrete profile burial history controls on the dia- formation, Saldanha Bay (South genetic evolution of reservoir sand- Africa). Sedimentology 24, 657– stones: evidence from the Lower 674. Cretaceous Serraria sandstones in Körner, F., Schneider, J. W., Ho- the Sergipe-Alagoas Basin. In: ernes, S., Gand, G., Kleeberg, R., Morad, S. (Ed.), Carbonate Ce- 2001. Climate and continental sed- mentation in Sandstones. Vol. 26 of imentation in the Permian of the Spec. Publ. Int. Assoc. Sediment., Lodève Basin (South France) - Oxford, pp. 107–140. sedimentology, geochemistry, cy- Gebhardt, U., 1988. Mikro- clostratigraphy. Terra Nos. 2001/4, faziesanalyse und stratigraphisch- 25–29. regionalgeologische Interpretation Lamplugh, G. W., 1902. Calcrete. terrestrischer Karbonate der varis- Geol. Mag. 9, 75. tischen Molasse (Mitteleuropa, Land, L. S., 1980. The isotopic Permokarbon). Freiberger Forsch.- and trace element geochemistry of H. C427, 30–59. dolomite; the state of the art. Soc. Goudie, A. S., 1973. Duricrusts in econ. Paleont. Miner., Spec. Publ. Tropical and Subtropical Land- 28, 87–110. scapes. Claredon Press, Oxford. Langbein, R., Peter, H., Schwahn, Goudie, A. S., 1983. Calcretes. In: H.-J., 1982. Karbonat- und Sulfat- Goudie, A. S., Pye, K. (Eds.), gesteine. VEB dt. Grundstoffind- Chemical sediments and geomor- ustrie, Leipzig. phology: precipitates and residua Lumsden, D. N., 1979. Discrepancy in the near-surface environment. between thin section and x-ray es-

19 timates of dolomites in limestone. carbonate content in sediments, J. sediment. Petrol. 49, 428–435. soils and other materials. N. Jb. Machette, M. N., 1985. Calcic Soils Mineral., Mh. 1971, 466–469. of the south-western United States. Netterberg, F., 1967. Some road mak- Geol. Soc. Amer., Spec. Pap. 203, ing properties of South African cal- 1–21. cretes. In: 4th Reg. Conf., Soil Mack, G. H., James, W. C., Monger, Mechan. Found. Eng. Vol. 1. Cape H. C., 1993. Classification of pale- Town, pp. 77–81. osols. Geol. Soc. Amer., Bull. 105, Netterberg, F., 1980. Geology of 129–136. southern African calcretes. I. Ter- Marriott, S. B., Wright, V. P., 1993. minology, description, macrofacies Palaeosols as indicators of geomor- and classification. Trans. Geol. phic stability in two Old Red Sand- Soc. S. Afric. 83, 255–283. stone alluvial suites, South Wales. Pagnier, H. J. M., v. Tongeren, P. J. geol. Soc. Lond. 150, 1109–1120. H. C., 1996. Upper Carboniferous Marriott, S. B., Wright, V. P., of borehole ’De Lutte 6’ and evalu- 1996. Sediment recycling on Siluro- ation of the Tubbergen Formation Devonian floodplains. J. geol. Soc. in the eastern and southern parts of Lond. 153, 661–664. the Netherlands. In: v. Tongeren, McCrea, J.-M., 1950. The isotope P. H. C. (Ed.), The ’De Lutte’ chemistry of carbonates and a Project. A multidisciplinary study palaeotemperature scale. J. Chem. of the Late Carboniferous in East Phys. 18, 849–857. Twente (the Netherlands). Vol. 55. McFadden, L. D., Tinsley, J. C., Meded. Rijks Geol. Dienst, pp. 3– 1985. Rate and depth of pedogenic- 30. carbonate accumulation in soils; Paul, J., 2004. Permokarbon am formulation and testing of a com- Kyffhäuser. in prep. . partment model. Geol. Soc. Amer. Pimentel, N. L., Wright, V. P., Spec. Pap. 203, 1–21. Azevedo, T. M., 1996. Distinguish- Morad, S., Al-Aasm, I. S., Longstaffe, ing early groundwater alteration F. J., Marfill, M., Ros, L. F. D., effects from pedogenesis in ancient Johansen, H., Marzo, M., 1995. alluvial basins: examples from the Diagenesis of a mixed siliciclas- Palaeogene of southern Portugal. tic/evaporitic sequence of the Mid- Sediment. Geol. 105, 1–10. dle Muschelkalk (Middle Triassic), Quast, A., 2003. Calcretes aus the Catalan Coastal Range, NE jungpaläozoischen Bodenbildun- Spain. Sedimentology 42, 749–768. gen: ein möglicher Proxy fur¨ die Morad, S., De Ros, L. F., Nystuen, CO2-Konzentration der Paläoat- J. P., Bergan, M., 1998. Carbonate mosphäre. Diss., Univ. Göttingen, diagenesis and porosity evolution http://webdoc.sub.gwdg.de/diss in sheet-flood sandstones: evidence /2003/quast/index.html. from the Middle and Lower Lunde Retallack, G. J., 1990. Soils of the Members (Triassic) in the Snorre past: an introduction to paleope- Field, Norwegian North Sea. In: dology. Harper Collins Academic, Morad, S. (Ed.), Carbonate Ce- London. mentation in Sandstones. Vol. 26 of Richter, D. K., 1974. Entstehung und Spec. Publ. Int. Assoc. Sediment., Diagenese der devonischen und pp. 53–85. permotriassischen Dolomite in der Muller,¨ G., Gastner, M., 1971. The Eifel. Contr. Sediment. 2, 1–101. ”Karbonat-Bombe”, a simple de- Richter, D. K., 1985. Die Dolomite vice for the determination of the der Evaporit- und der Dolocrete-

20 Playasequenz im mittleren Keuper dependence of oxygen isotope frac- bei Coburg (NE-Bayern). N. Jb. tionation of CO2 from magnesite- Geol. Paläont., Abh. 170, 87–128. phosphoric acid reaction. Geochim. Rosenbaum, J., Sheppard, S. Cosmochim. Acta 60, 589–593. M. F., 1986. An isotopic study of Soil Survey Stuff (Ed.), 1975. Soil siderites, dolomites and ankerites Taxonomy: a basic system of soil at high temperatures. Geochim. classification for making and inter- Cosmochim. Acta 56, 1147–1150. preting soil surveys. Vol. 436 of Hb. Royer, D. L., Berner, R. A., Beer- U.S. Dep. Agricult. ling, D. J., 2001. Phanerozoic at- Spötl, C., Wright, V. P., 1992. mospheric CO2 change: evaluat- Groundwater dolocretes from the ing geochemical and paleobiologi- Upper Triassic of the Paris Basin, cal approaches. Earth-Sci. Rev. 54, France: A case study of an arid, 349–392. continental diagenetic facies. Sedi- Saigal, G. C., Bjørlykke, K., 1987. mentology 36, 1119–1136. Carbonate cements in clastic reser- Trewin, N. H., 1987. Devonian of St. voir rocks from offshore Norway Cyrus and Milton Ness. In: Trewin, - relationships between isotopic N. H., Kneller, B. C., Gillen, C. composition, textural development (Eds.), Excursion guide to the ge- and burial depth. In: Marshall, ology of the Aberdeen Area. Geol. J. D. (Ed.), Diagenesis of sedimen- Soc. Aberdeen, pp. 251–258. tary sequences. Vol. 36 of Spec. Trewin, N. H., Kneller, B. C., 1987. Publ. Geol. Soc. London, pp. 313– The Lower Old Red Sandstone of 324. New Aberdour. In: Trewin, N. H., Saigal, G. C., Walton, E. K., 1988. Kneller, B. C., Gillen, C. (Eds.), On the occurence of displacive cal- Excursion guide to the geology of cite in Lower Old Red Sandstone the Aberdeen Area. Geol. Soc., Ab- of Carnoustie, eastern Scotland. J. erdeen, pp. 131–142. sed.. Petrol. 58, 131–135. Watts, N. L., 1980. Quaternary pe- Schneider, J. W., Gebhardt, U., 1993. dogenic calcretes from the Kala- Litho- und Biofaziesmuster in hari (southern Africa. Sedimentol- intra- und extramontanen Senken ogy 27, 661–686. des Rotliegend (Perm, Nord- und Wieder, M., Yaalon, D. H., 1982. Mi- Ostdeutschland). Geol. Jb. A131, cromorphological fabrics and de- 57–98. velopmental stages of carbonate Schneider, J. W., Rößler, R., 1995. nodular forms related to soil char- Permische Calcisol-Paläoböden acteristics. Geoderma 28, 203– 220. mit Rizolithen und Wirbeltier- Worden, R. H., Matray, J. M., 1998. resten - Sedimentation, Lebe- Carbonate cement in the Trias- welt und Klimaentwicklung im sic Chaunoy Formation of the Rotliegend der Härtensdorf- Paris Basin: distribution and effect Formation. Veröff. Mus. Nat.-Kde. on flow properties. In: Morad, S. Chemnitz 18, 53–70. (Ed.), Carbonate Cementation in Scholle, P. (Ed.), 1979. A Color Illus- Sandstones. Vol. 26 of Spec. Publ. trated Guide To Constituents, Tex- Int. Assoc. Sediment., pp. 163–178. tures, Cements and Porosities of Wright, V. P., Peeters, C., 1990. Sandstones and Associated Rocks. Origins of some early Carbonif- Vol. 28 of Mem. Amer. Assoc. erous calcrete fabrics revealed by petrol. Geol. cathodoluminescence: implications Sharma, G. S. D., Patil, D. J., for interpreting the sites of cal- Gopalan, K., 2002. Temperature crete formation. Sediment. Geol.

21 65, 345–353. Wright, V. P., Tucker, M. E., 1991. Calcretes: an introduction. In: Wright, V. P., Tucker, M. E. (Eds.), Calcretes. Vol. 2 of Repr. Ser. In- tern. Assoc. Sediment., pp. 1–22. Wright, V. P., Turner, M. S., An- drews, J. E., Spiro, B., 1993. Mor- phology and signiﬁcance of super- mature calcretes from the upper Old Red Sandstone of Scotland. J. geol. Soc. Lond. 150, 871–883. Yaalon, D. H., Kalmar, D., 1978. Dy- namics of cracking and swelling clay soils: Displacement of skeletal grains, optimum depth of slicken- sides, and rate of intra-pedogenic turbation. Earth Surface Proc. 3, 31–42.