PALEOPEDOLOGY - a Geologist’s reflections on the Third International Symposium and Field Workshop on Paleopedology in Rauischholzhausen, Germany, 1997

Posted by Prof. Andrea Mindszenty 15 Jun 1999 13:46:25

Department of Applied and Environmental Geology, Faculty of Soil Science, Eotvos University, Budapest, Hungary

Ancient soils

Soils of the past may be studied with widely different background and for various reasons. Modern soil scientists when trying to understand
the response of various substrata to actual climatic and topographic conditions, are often inhibited in doing so be anomalies caused by relict or
exhumed soils blanketing the actual landsurface but not related to the actual climate. For them, the study of ancient (but not necessarily very
old) soils is necessary to comprehend the anomalies observed. Geologists , when studying sedimentary rocks, often encounter buried and
fossilized soil horizons or even complete soil profiles which by their shear presence are useful indicators of subaerial exposure. Also, such
fossil soils may provide information about past climatic (and/or topographic) conditions. One thing should be kept in mind, however, namely
that the environmental information provided by ancient soils (be it relict, buried, exhumed or fossil) will never be as direct and unequivocal as it
may be in the case of modern soils fully in equilibrium with the actual environment. Ancient soil profiles are very rarely complete phisically and
very rarely monogenetic geochemically mineralogically and texturally. What is more, ancient soils encountered in the stratigraphic column will
invariably have been subject to the rpocesses of diagenesis - familiar to the geologist but unusual to scientists of modern soils (just as many
biologically mediated elementary processes are less well known for the geolodists).

Paleopedology - soil science or geology?

Very clearly paleopedology is a discipline halfway between pedology and sedimentary geology and scholars coming from either side can not
avoid getting acquainted at least with the essentials of the other side. Under “essentials” not only methods and subdisciplines (like
microbiology, botanics, mineralogy, geomorphology or basic structural geology) shold be consedered but rather the approach - a quality
extremely difficult to define. The approach of a scientists is a cumulative property acquired normally during hai basic training, when he is
comtinuously surrounded by a particular scientific atmosphere created by his professors (for him the representatives of his chosen discipline).
The geological approach, for example, enables the scholar to look at all natural phenomena (be it mineralogical, biological or chemical) with a
time/space framework in mind. He combines the conventional three-dimensional view of natural sciences (like geomorphology) with the
historical perspective< with time as a fourth dimension, much like archeologists do.

Althought time - manifested as the duration of pedogenesis - is a parameter realized also by modern soil scientists, their concept of a “long
time” is a radically different from that of the geologist. For most soil scientists one million year would be a vary long time, whereas for the
geologists dealing with hundreds of million years as a routine, one million year would be an “instant”, and most geologists working with
pre-Quaternary profiles would be extremely happy when having data sets with a time-resolution of one million year. Quite naturally, it is an
over-simplification to say that the difference between the approach of the two groups would be resulted only by the different time concepts.
There are additional differences rooted in basic training, in the organization of curricula, etc. The time concept, however, is crucial in
determining the approach.

The Meeting

The Meeting in Rauischholzhausen provided excellent opportunity for people of different approach to interact, exchange ideas and make more
than one important step toward the establishment of a common language - the prerequisite of mutual understanding. That the progress was not
quite as big as it might have been was by no means the foult of those present, but rather of those whom we greatly missed. Over 90% of the
participants were soil scientists having modern pedology as their basic training (and specizlized in Holocene to Pleistocene soils). Geologists
interested in older soils were in minority. One reason for that was obviously the unfortunate almost-overlap of the Rauischholzhausen
workshop with the European Meeting of IAS (the International Associztion of Sedimentologists) in Heidelberg (just two weeks before
Rauischholzhausen). Two meetings, as closed as temporally, often discourages participation at one or the other. They should be either
consecutive, or much farther apart. Probably this is why many sedimentologists trained in paleopedology have not come, though we all could
have had greatly benefited from their presence.

Still, those present, have their best to bridge the gap between the two disciplines, and the effort was undoubtedly worthwhile. We, geologists
have learned a great deal from our fellow soil scientists and the struggle to understand the other side and to make ourselves understood has
proved to be an invaluable help in clarifuing our ideas about paleopedology in general and about our older-than-Pleistocene soils in particular.
For the opportunity to achieve this, sincere thanks should go to all the organizers and among them particularly to P.Felix-Henningsen and
Thomas Scholten Special thanks are due to Drs a.Semmel, M.Weidenfeller and L.Zoller, that guides of the field-trips in the Rheingraben and
surrounding.

About the Paleopedology Glossary

The best Glossaries are always open to additions. The 2nd draft distributed in Rauischholzhausen is certainly a thoughtful and an almost
perfectly concise one, so that it is hard to come up with any additions. The more so that it was apparently preceeded by long brain-storming
sessions of a community of soil scientists far more authentic in the matter of terminology than an average geologist would be.

After these intriductory remarks let me take the liberty and comment on some of the items included in it.

Geosols

In order to clarify the difference between a geosol and a paleosol perhaps it could be emphasized, that the term geosol refers to an assemblage
of ancient soils rather then to any particular paleosol. The geosol is “...a whole soilscape that can be recognized as a laterally extensive
stratigraphic horizon”. (Retallack 1990, Follmer 1978, Morrison 1978 etc.)

In addition ti the terms cumulic and compround perhaps also composite should be mentioned. This latter term was used by Wright and Marriot
(1995) to describe the situation referred to as “Lateral merging af several geosols” in the Glossary.

Pedofacies

A reference to terms like catena, toposequence and chronosequence would be useful here.

Polygenetic soil

The last sentence of this entry (“Minor episodes of deposition may have occured between or within the periods of soil development, but were
not sufficient to leave layers of unaltered material”) seems to be simewhat out of place here. It could be added rather to the entry on geosols,
where the term cumulic soil is explained.

Soil

Implict in the present definition it is included that sediments modified by pedogenic precesses to an extent which permits the recognition of the
aleration also in the lithified sedimentary rock (e.g. fossilized root molds and/or associated primitive gley spots, etc.) but not displaying distinct
horizontation (like weakly developed entisols for example) would not be considered as soil. The stratigraphic record developed counterparts
would be the sources of valuable information for the geologist looking for the record of episodes of subaerial exposure within a shallow-marine
or a fluvial sedimentary rock suite. In a study focusing to such paleosol sequences it would be very inconvenient to exclude these weakly
developed soils from under the term soil, so it is highly recommended to adjust the definition (in the paleopedological context) to meet also the
requirement of geology. The proposed new sentence would read as follows: “a three-dimentional body on the surface of the earth composed of
mineral and/or arganic material, air and water, and formed by the impact of environmental factors acting on parent materials over a period of time
to produce recognizable alteration of the parent material in some times resulting in the establishment of a sequence of horizons”.

Soil sediment

As a synonym the term pedorelict (Retallack 1983, 1990) should be indicated. Also, reference to Ida Valeton’s Laterite Derived Facies (LDF)
could be made here (Valeton 1983 and Valeton & Wilke 1993).

Vetusol

This is a very loose category. Also some of the statements included in the definition may not be of overall validity.

Soils occuring a major regional unconformities (the geological term for long-lasting subaerial episodes recognized in the stratigraphic record)
are very rarely “monogenetic” (if not rof other reasons, at least because of the addition of aeolian dust from sources other than their parent
rock!). In fact the overwhelming majority of geosols recognized by geologists are known to be poligenetic in origin.

In the case of soils formed over very long periods of time, the absence of “detectably different assemblages of soils features” is very probably
not the result of “landsurface stability” or of the “small variability of environmental factors” but rather of the inertia of the soil formed during
earlier stages of the long history of the subaerial exposure phase. Certain well-developed soils (like oxisols or ferrasols) may not have the
peogenetic potential to reacr to high-frequency climate changes having a duration of 20.000 years. As an example the case of Mesozoic shallow
water carbonate sediments, and their intercalated bouxites (presumably representing several million years of exposure) could be cited (see also
D’Argenio & Mindszenty 1995). Carbonate depositional environments being very sensitive to environmental changes were shown to record
high frequency climate changes wihthin the intertropical belt (Fischer 1964, 1991; Goldhammer et al.1990, D’Argenio et al 1992 etc.). Such high
frequency cyclicity detected in carbonate rocks of a given locality indicate that high frequency changes did take place in that area in the past.
Bauxites (equivalents of oxisols) intercalated in the sequence generally do not record such high frequency changes, which does not necessarily
mean that for several million years high frequency changes would have been stopped, Rather it means that the soil which during a number of
those changes finally has riched the stage of an oxisol, was unable to reflect any more changes and gives the false impression of unchanged
conditions throughout the period studied!

By now means the definition should suggest restrictions in time, like the remark in the last sentence referring to Quaternary climate changes.
Climate changes as significant as the Quaternary ones have taken place several tmes during the Earths’ history, so there is no reason for the
discrimination.

With the above considerations in mind the definition of vetusols could be changed as follows:

“soils formed over a very long period of time (order of magnitude > 105 years) during which (a) either the influence of pedogenetic and eventual
sedimentary processes (like repeated reworking) has resulted in an assemblage of often polygenetic soil features which may not permit to
inequivocally identify them as belonging to one or the other of the categories of Soil Taxonomy. In some cases, however, they may be identified
as oxisols or ultisols or their partly resedimented rpoducts (like the LDF of I.Valeton).; or (b) vetusols may be produced also by long lasting
conditions of stable land surface and minor or no variations of environmental factors. Most vetusols occur on old geomorphic surfaces in
tropical and subtropical regions where the influence of high frequency climate changes was minor.”

Classification of paleosols

Efforts to elaborate a reasonable classification of paleosols have been numerous. Following the first arttemps of Polynov (1927) whose four
categories (secondary soils, two-stage soils, fossil soils and ancient weathering products) even if rather straightforward, were essentialy
genetic categories, there was a general tendency towards a more phenomenological ckassification utilizing more and more directly the
terminology of modern soil taxonomy of which the best examples are the classifications of Retallack (1990) and Mack et al (1993). The former
simply adopted all the categories of the US Soil taxonomy for paleosols while the latter have modified it an ended up with only 9 paleosol
clsses. This approach has two inherent shortcomings: (a) every attempt to categorize a given paleosol would necessarily be biased by the fact
that most criteria used in modern soil science (like ion exchange capacity, soil moisture regime, etc.) instead of being directly measured, will
have to be “guessed” on the basis of their supposed effects on the soil. (b) after having identified the paleosol as the fossil equivalent of its
modern counterpart by using the above “guessed” criteria, the result will be highly temptating to be used for an analogy-based environmental
reconstruction. Mental exrcises like this would obviously violate the rules of formal logic (conclusion used as a premiss in the same context).

May be a simple phenomenilogical classification, based in geological rather than pedological criteria and combined with the interpretation of the
effects of those elementary pedogenic processes which do have a high enough preservation potential to survive throughout diagenesis and
perhaps even incipient metamorphism would be the right solution.

To elaborate a geological classification, paleosols should be treated as sedimentary rock suites. The relationship of any sedimentary rock body
to the sediments it is enclosed by is one of the crucial criteria of such a classification. Being purely phenomenological, this criterion has also the
benefit of being very easy to convert to genetical terms.

With such an approach we may say that there are principally two kinds of paleosols

(a) Paleosols associated with major regional unconformities

(b) Paleosols associated with international unconformities

(a) is obviously something more or less equivalent to vetusols in the Glossary. It may be added, that most major subaerial unconformities are
associated with mature paleosols like e.g. laterites, bauxites, terra rossa, silcretes and xertain calcretes. They may be partly or entirely
resedimented, and may contain abudant pedorelics. These paleosols do not continually or uniformly cover quite all the former subaerial
landscape, because when, after a break, sedimentation is resumed again, erosion may destroy part of the soils formed during the break. Also
they may form paleo-catenae reflecting the underlying paleorelief and therefore useful in the paleotopographic reconstruction.

(b) Paleosols may occur in every sedimentary environment where deposition is more or less regularly interrupted by shorter or longer episodes
of non-deposition and suaerial exposure. Environments like this are e.g. most aeolian and alluvial settings, delta-platforms, volcano sedimentary
terrains and shallow-water carbonate platforms. Depending on the rate at which sedimentation resumes after an episode of interruption, these
soils may be singles (=compounds), composites (=nerging) and cumulic (=cumulates, soil-sediment complexes) (c.f. Wright & Marriott 1996).

Intraformational paleosols are substantially different from those associated with major regional unconformities, not only because of the
generally much longer duration of pedogenesis of the latter. At major regional unconformities the parent material, i.e. the substratum of soil
formation, is generally a mature rock surface consisting of sediments already passed through all stages on near-surface and burial diagenesis,
sometimes even of metamorphism. Their pedogenetic alteration can thus be considered as an epigenetic or telogenetic process. On the
contrary, intraformational (or synsedimentary) paleosols often form on the surface of unconsolidated, sometimes even loose sediments, i.e.
pedogenetic alteration in this case is essentially a diagenetic process taking place under meteoric conditions. In the first case, the beginnings
of pedogenesis are necessarily restricted to joints and fissures, where plants may penetrate the hard rock substrate and the saprolite consists of
fragmented hard rock displaying a distinct porosity increase as compared to the parent material. Intraformational pedogenesis, on the other
hand, attacking the fresh, uncompacted sediment, penetrates the substratum along primary pores and forms a diffuse weathering front along
which sometimes, due to early cementation, primary porosity is in fact plugged instead of being increased (cf.calcretes!). With time such
cementative phenomena may result in lithification of the originally unconsolidated sediment and thus in a gradual change of the
intraformational regime into that known from situations at major regional unconformities, i.e. early and late (intraformational and telogenetic)
pedogenesis may overlap.

This simple twofold division is easy to combine with any one of the previously suggested detailed classifications (depending on the amount of
information avaolable). Classifications with too many categories should be discouraged just as the forced analogy to modern soils is. The study
of as many soil features as possible and their simple interpetation in terms of elementary soil forming processes would probably be the most
fruitful approach - at least when paleopedological research is intended to contribute to the paleoenvironmental reconstruction.

What else?

Further collaboration of soil scientists and geologists to facilitate a betterunderstanding of past soil forming environments is highly desirable:

Analogy-based interpretation of fossil soil features observed in older-than-Pleistocene paleosols would be easier when geologists dealing with
such old soils would have appropriate chronosequences and/or climosequences available from Pleistocene/Holocene settings studied by soil
scientists.

The correct identification of paleosols as approximate equivalents of one or the other of modern soil taxa could be facilitated when the
preservation potential of diagnostic soil features were known. Research focussed to improve our understanding of the fossilization of
diagnostic soil features and the organization of further interdisciplinary workshop to discuss these problems should be encouraged.

References cited

D’Argenio, B.-Ferreri, V.-Iorio, M.-Longo, G.-Raspini, A. (1992): Early Crateceous eustatic oscillation under astronomical forcing and relkated
problems in time calibration. The case of Southern Italy. In: CNRS-IFP Sequence Stratigraphy of European Basins, 406-407, Dijon, Abs.Vol.

D’Argenio, B.-Mindszenty, A. (1995): Bauxites and related paleokarst: Tectonic and climatic event markers at regional unconformities. Eclogae
Geol.Helv. 88.3: 453-499

Fischer, A.G. (1964): The Lofercyclothems of the Alpine Trassic. Bull.Kans.State Geol.Surv., 169: 107-149

Fischer, A.G. (1991): Orbital Cyclicity in Mesozoic strata. In Einsele, G.-Ricken, W.-Seilacher, A. (eds.): Cycles and Events in Stratigraphy.
Springer Vrlg. Berlin, 48-22

Follmer, L.R. (1978): The Sangamon soil on its type erea - a review. In Quaternary soils, W.C.Mahaney (ed.), 125-65. Norwich: Geoabstracts

Goldhammer,, R.K.-Dunn, P.A.-Hardie, L.A. (1990): Depositional cycles, composite sea-level changes, cycle stacking pattern and the hierachy
of stratigraphic forcung. Examples from Alpine Triassic carbonates. GSA Bull. 102: 435-562

Mack, G.H.-James, C.W.-Monger, C.H. (1993): Classificationof paleosols. GSA Bull. 104: 129-136

Morrison, R.B. (1978): Quaternary soil stratigraphy - concepts, methods and problems. In Quaternary soils, W.C.Mahaney (ed.), 77-108,
Norwich: Geoabstracts

Polynov, B.B. (1927): Contributions of Russian scientists to paleopedology. Leningrad: USSR Academy of Sciences (as cited by Retallack
1990)

Retallack G.J. (1990): Soils of the Past Allen & Unwin (Australia) Ltd. pp 1-520

Retallack, G.J. (1983): A paleopedological approach to the interpretation sedimentary rocks: the mid-Tertiary fossil soils of Badlands National
park, South Dakota. GSA Bull. 94: 824-840

Valeton, I. (1983): Klimaperioden lateritischer Verwitterung und ihr Abbild in den Synchronen Sedimentationsraumen. Z.dt.Geol.Ges. 134:
413-452

Valeton, I.-Wilke, F. (1993): Tertiary bauxites in subsidence ereas and associated laterite derived sediments in northwestern India. In
Fuchtbauer H.-Lisitzin, A.P.-Milliman, J.D.(eds.). Contributions of Sedimentology No.18. Scweizerbart’sche Vrlg., Stuttgart

Wright, P.V.-Marriott, S.B. (1996): A quantative approach to soil occurrence in alluvial deposits and its application to the Old Red Sandstone
on Britain. Journ.Geol.Soc.London, 153: 907-913.