Earth-Science Reviews 103 (2010) 122–134

Contents lists available at ScienceDirect

Earth-Science Reviews
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r s c i r ev

Soil science applications in archaeological contexts: A review of key challenges

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

Article history: The recent emergence and application of Earth Science techniques, such as elemental analysis, to detect
Received 4 December 2008 isotopes, biomarkers, trace and ultra trace metals, in combination with long established techniques like
Accepted 7 September 2010 magnetic susceptibility and micromorphology, can allow fascinating insights into the analysis of soils at
Available online 17 September 2010
archaeological sites. Soil studies can reveal how humans in prehistory used the landscape and defined space
through their activities. However, these new approaches do not wholly address persistent problems
Keywords:
pedoarchaeology
associated with making inferences about past human activity from soils. These challenges include:
soil processes equifinality; distinguishing property–process relationships; identifying anthropogenic soil processes; the
pedology interdependency of the soil forming factors; and difficulties with soil dating. This paper reviews more than a
equifinality decade of pedoarchaeological studies, structured around new approaches to addressing these challenges.
The paper outlines a staged framework which helps to create a systematic interpretation of soil processes and
properties, and considers the impact of anthropic soil processes and properties in this context. This approach
can be used as a guide to ensure that a rigorous and reproducible approach is taken to the study of soils at
archaeological sites. In making this framework explicit, the paper finds that establishing property–process
relationships in the soil is an essential precursor to reliable pedoarchaeological interpretation. It is argued that
in the future, new applications developed in the Earth Sciences that aid our understanding of archaeological
soil processes in three dimensions, will be able to contribute the most to addressing persistent challenges in
pedoarchaeological interpretation.
© 2010 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
1.1. The soil–sediment continuum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
1.2. Soil formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
1.3. Paleosols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
1.4. The soil archive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
1.5. Approaches to the study of archaeological soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
2. The five major challenges to soil study at archaeological sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
2.1. Equifinality: the challenge to identifying discrete process–response relationships . . . . . . . . . . . . . . . . . . . . . . . . . . 125
2.1.1. Thin section analysis (soil micromorphology) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
2.1.2. Experimental pedology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
2.1.3. Historical sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
2.1.4. Elemental analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
2.1.5. Biomarkers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
2.2. Distinguishing property–process links where preservation is poor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
2.2.1. Bioturbation and site integrity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
2.3. The challenge of isolating anthropogenic soil processes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
2.3.1. Geophysical approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
2.3.2. Activity areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
2.4. Difficulties of soil interpretation due to the interdependency of the soil forming factors . . . . . . . . . . . . . . . . . . . . . . . 129
2.4.1. Accelerated soil erosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
2.5. Difficulties with soil dating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
2.6. Summary of challenges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

⁎ Tel.: +44 1865 483185; fax: +44 1865 483937.

0012-8252/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.earscirev.2010.09.002
 H. Walkington / Earth-Science Reviews 103 (2010) 122–134 123

3. A framework for interpreting soils at archaeological sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

1. Introduction and slope processes. Sediments may have pedogenic (soil forming)
processes acting upon them, and this blurs the distinction. The
Despite the numerous opportunities which the study of soils at possibility of interaction between sedimentary and pedogenic
archaeological sites provide, as well as the wealth of scientific processes is further complicated by anthropogenic impact. Separating
approaches available for soil analysis, there remain several persistent these three influences at archaeological sites can be challenging. In
challenges associated with making inferences about past human order to do so it is important to understand the fundamental
activity from soils. The recent emergence and application of earth characteristics of soils which have not been directly influenced by
science techniques for the analysis of soils at archaeological sites has human activity.
generated new levels of understanding of human activities and use of
the landscape. Inductively coupled plasma atomic emission spectros- 1.2. Soil formation
copy (ICP-AES) and inductively coupled plasma mass spectrometry
(ICP-MS), can help to detect isotopes, lipids and biomarkers and trace Soils have traditionally been studied through the framework of the
and ultra trace metals. In combination with long established soil forming factors, i.e., climate, organisms (flora, fauna and humans),
techniques like magnetic susceptibility, phosphate analysis and relief, parent material and time (Jenny, 1941). Soil development
micromorphology, the information gained can allow fascinating therefore follows a pathway related to the interaction of these factors
insights into the analysis of soils at archaeological sites. Soil studies in a geographical location. The relative importance of temperature
can reveal how humans in prehistory used the landscape and defined and precipitation patterns; floral, faunal and human impact; topo-
space through their activities. However, these new techniques do not graphic position and drainage; the rock or sediment in which the soil
wholly address several persistent problems associated with making develops and the time it has been forming varies significantly.
inferences about past human activity from soils. However, more recently there has been greater recognition in the
This paper begins by briefly describing soils and sediments and literature of an approach termed ‘dynamic denudation’ (Johnson,
their differences. It considers frameworks for understanding soil 2002; Johnson et al., 2005a,b). This approach stresses the importance
development and outlines how paleosols can be regarded as archives. of biodynamic processes, particularly bioturbation. Bioturbation is
It goes on to review five major challenges when studying soils at crucial in developing soil thickness relationships, i.e., the upbuilding,
archaeological sites. These challenges include equifinality; distin- deepening and removal process pathways of soil formation. Johnson
guishing property–process relationships; identifying anthropogenic et al. (2005a) argue that biodynamic processes contribute to soil
soil processes; the interdependency of the soil forming factors; and formation to a much greater extent than previously thought. Related
difficulties with soil dating. Each challenge is exemplified using recent to the notion of soil deepening, Tanadarich et al. (2002) argue that
literature relating to a specific soil related theme, such as the there is a tension between the geological concept of soil depth (the
anthropic origin of Terra preta soils from the Amazon, soil erosion full depth of weathering) and the shallower pedological concept (the
in the eastern Mediterranean, evidence of cultivation and manuring solum) which traditionally addressed the needs of agriculture. Instead
land management practices in north west Europe, the identification of they use the concept of a ‘pedoweathering profile’ so that subsolum
activity areas in Meso America, site integrity and bioturbation at properties cannot be ignored. Indeed, by investigating the depth of
sandy sites in the US. Finally the paper outlines the interpretive just one soil process, clay translocation, Johnson et al. (2003) noted
framework currently adopted in pedoarchaeology. that illuvial clay can be deposited at great depths in the regolith and at
the contact with bedrock. The pedoweathering profile concept takes
1.1. The soil–sediment continuum into account such depths which traditional definitions of soil do not.
The depth of burial of an archaeologically important soil can be
Soils and sediments are composed of similar components, and instrumental in helping it to preserve information about the past, but
form a continuum over the landscape. The point at which a sediment other environmental factors related to climatic, biotic, topographic
becomes a soil is related to the vertical patterning of properties, that and parent material characteristics also influence the information that
cannot arise from sedimentation alone, created by in situ transforma- can be gleaned from ancient landsurfaces.
tions at the earth's surface. Sediments comprise layered, unconsoli-
dated materials of lithic or organic origin. They usually show 1.3. Paleosols
stratigraphy, are paler and have a lower organic content than soils.
Although they can show iron staining, they do not have the Despite a variety of definitions, paleosols still remain best defined
characteristic weathering horizons that are found in soils. Sediments as soils formed in landscapes of the past (Yaalon, 1986). Paleosols can
tend to accumulate during phases of instability and can be deposited be classified according to their state of preservation. Relict paleosols
very rapidly (Waters, 1992; Rapp and Hill, 1998; Goldberg and are those soils formed on pre-existing landscapes, under previous
Macphail, 2006). environmental conditions, but which have always remained at the
Soil formation (pedogenesis) occurs in situ, to create a continuous surface. Because they have not been buried, they exhibit horizons or
layer with weathering horizons such that the chemical and textural features from previous environments, over printed by modern
composition changes in line with earth surface processes (Waters, processes. Their features are therefore polygenetic (Goldberg and
1992; Rapp and Hill, 1998; Goldberg and Macphail, 2006). Soils Macphail, 2006). Buried paleosols are the most useful for archae-
mature slowly, thereby developing a sequence of recognisable ologists because they were formed on a landscape of the past but
horizons which are genetically linked to each other. They form over buried by younger sediment such as loess, coversand, till, alluvium
whole landsurfaces during periods of landscape stability. They (Catt, 1987) or even by anthropogenic activity, such as construction of
therefore develop catenary sequences due to differences in drainage earthworks (French, 2003). For some classifications, buried soils
124 H. Walkington / Earth-Science Reviews 103 (2010) 122–134

should be covered by sufficient material for them to be isolated from 4) Soils can record human use and management of a landscape. In
contemporary processes (Birkeland, 1984). In contrast, Schaetzl and addition to retaining artefacts, some soils may preserve evidence of
Anderson (2005) argue that a soil is buried when the overlying human impacts e.g. manuring (Sandor, 1992), managing livestock
sediment is not genetically related to the paleosol, in contrast to (Heathcote, 2002), monument construction (Macphail, 1987) and
approaches relying on measured depths (Soil Survey Staff, 2010). activity areas (Simpson et al., 1998b; Middleton, 2004; Shahack-Gross
Undisputed identification of a buried soil is via a palaeocatena et al., 2005; Cook et al., 2006; Hutson and Terry, 2006; Hutson et al.,
(Valentine and Dalrymple, 1976b); this means that a whole landsur- 2007) which can be used to build a picture of past human use of space.
face has been buried, revealing a catenary sequence of soils which
differ according to their slope position. Where the overlying sediment 1.5. Approaches to the study of archaeological soils
is removed, a buried soil is termed an exhumed paleosol (Waters,
1992; French, 2003; Schaetzl and Anderson, 2005; Goldberg and The application of chemical signature techniques, such as
Macphail, 2006). elemental analysis, to archaeological settings has revolutionised the
way in which evidence of human activity can be inferred from the soil.
However, the presence or absence of elements alone may be
1.4. The soil archive inadequate to effectively understand the relationship between soil
properties and pedogenic, diagenetic and/or anthropogenic processes.
Soil is an archive that can be used to interpret archaeological As a result, multimethod approaches are becoming standard practice
deposits (Mandel and Bettis, 2001). Soils have great archaeological in pedoarchaeology. Soil micromorphology is a technique which is
potential, as they form during periods of landscape stability, and can very commonly employed in conjunction with other methods. In 1938
therefore be used to infer physical and chemical conditions which Kubiena introduced soil microscope analysis to soil scientists
were present over significant periods of time. Indeed, soils may be the (Nortcliff, 1985), from which new sampling techniques have been
only local terrestrial environmental archive available. The biotic, developed (for example, Goldberg and Macphail, 2003; Josephs and
climatic, geomorphological and diagenetic processes which operate Bettis, 2003). Despite the time consuming nature of the preparation
on archaeological soils affect their preservation potential (Jans et al., and analysis of thin sections, they can still provide the crucial spatial
2002; Scott-Jackson and Walkington, 2005). (visual and morphological) framework with which chemical signa-
Pedology is the study of soils as naturally occurring phenomena, tures can be associated. This type of research can be conducted while
taking into account their composition, distribution and processes of the three dimensional integrity of the soil remains intact, for instance
formation. Pedoarchaeology brings together the disciplines of pedol- by describing soil properties in the field, or removing stabilised soil
ogy and archaeology in order that they can contribute to an blocks with orientation data recorded on the sample. This is
understanding of the soil processes which were operating at a site particularly useful in relation to archaeological work because if
when artefacts were deposited, and any subsequent processes which artefacts are present they remain in context, and the relationship
have occurred. It allows the examination of the archaeological between pedogenic and anthropogenic processes can be explored in
assemblage, including the unique set of properties at a site (artefacts, detail. Image analysis has been used in soil science to rapidly quantify
lithology, texture, structure, organic remains, etc.) in the context of visible patterns, such as pore space (Lima et al., 2006; Marcelino et al.,
the study of pedogenesis, where the relationship between soil 2006). For archaeological applications image analysis requires careful
properties has a genetic connotation. Pedoarchaeology integrates an interpretation, and should be used in conjunction with micromor-
understanding of the processes which formed the archaeological phology and other standard tests (Bryant and Davidson, 1996;
record at a site and which continue to affect it. Matthews et al., 1997; Goldberg and Macphail, 2006).
Soil processes can contribute to an understanding of archaeolog- The archaeological potential of a soil hinges on the ability to
ical sites in particular because the soil can act as an archive in a variety establish the relationship between soil properties and the processes
of ways. These include: which formed them. It is only once the spatial pattern has been
characterised that it is possible to understand the processes which
1) Soils represent process–response systems through the unique have formed a particular soil feature (Dalrymple and Theocharopou-
interaction of the soil forming factors at a particular site. Soils form los, 1987). Simonson first recognised that while all processes are
part of a landscape development archive. For example, the present in all soils, it is the balance of processes and interaction
presence of preserved sub surface tree throw hollows in the soil between them over time which gives the soil its spatial expression in
can provide evidence of previous vegetation cover (Langohr, the form of resultant soil properties (Simonson, 1978). Four broad
1993). In some areas, paleosols can be traced across whole process groupings were recognised: additions, removals, transloca-
landscapes, creating soil stratigraphic units, such as the Sangamon tions (moved within the soil) and transformations (e.g. in situ change
soil in the USA (Hall, 1999). Buried palaeocatenas can preserve the such as chemical weathering) (Simonson, 1959). However, the
final phase of landscape development prior to burial (Valentine variance in preservation of soil features at archaeological sites will
and Dalrymple, 1976a,b). not necessarily reflect the balance of processes during their formation.
2) Soils reflect soil processes occurring both vertically and laterally. Properties with the best preservation can appear more dominant in
As three dimensional bodies, soils enhance our understanding of the soil record. Absence of a soil property does not mean that it is safe
the whole context of archaeological sites, through their link to to assume that a particular soil process did not occur. Instead an
taphonomic and geomorphological processes, i.e. through pedo- understanding of soil processes should give clues as to the associa-
geomorphology (Conacher and Dalrymple, 1977; Gerrard, 1992; tions of properties and suites of processes which operate in particular
Brooks, 2003). soil types, landscapes and climatic regimes.
3) Soils can store palaeoecological indicators such as pollen, phyto- This paper now reviews over a decade of soil related studies at
liths, coleoptera, mammal bones, etc. and are thus archives of archaeological sites in order to explore the five challenges to
environmental change (Lowe and Walker, 2006). Soils are a pedoarchaeological interpretation in turn.
widespread geoarchive, even when transport has occurred, the
resulting colluvial soils can still store environmental information 2. The five major challenges to soil study at archaeological sites
(Leopold and Völkel, 2007). Studies of soil development enable
researchers to contribute to an understanding of site formation The key challenges to soil study at archaeological sites are
and taphonomy (Leigh, 1998; Van Nest, 2002). interrelated, but are discussed separately for clarity. Each section begins
 H. Walkington / Earth-Science Reviews 103 (2010) 122–134 125

with an outline of an illustrative desired archaeological requirement Micromorphology is ideal for the production of feature microstrati-
from the soil, followed by an explanation of the pedological challenge, graphies (Cruise and Macphail, 2000), process–environment infer-
and a review of contemporary work to overcome the problem. ences in buried soils, and the sequencing of relict features in non
buried soils in order to reconstruct the depositional and post
2.1. Equifinality: the challenge to identifying discrete process–response depositional history of a site.
relationships
2.1.2. Experimental pedology
A process–response relationship is an identified causal link Methods for isolating soil processes have been developed through
between a soil process and the resultant soil property. Some soil experimental pedology in the laboratory. Using real soils Akamigbo
processes can result in such distinct and quantifiable properties, that and Dalrymple (1985) added soil clay suspensions in the laboratory to
features are regarded as diagnostic of a specific process, as in the case examine the pathways for the movement, and deposition, of clay sized
of clay illuviation (Soil Survey Staff, 2010: 10; Dalrymple and minerals. They made thin sections of the soils after each experimen-
Theocharopoulos, 1987). A central pedoarchaeological aim is therefore tally controlled illuvial event, in order to describe the morphological
to be able to link one diagnostic soil property to a single causal process response. Dalrymple and Theocharopoulos (1987) and Theocharo-
or suite of processes. A further aim is to quantify this relationship. poulos and Dalrymple (1987) conducted similar experiments with
In archaeological soils, the possible interaction of pedogenic, synthetic sponges. Their results showed that as well as allowing the
diagenetic and anthropogenic processes makes identifying property– accumulation of oriented clays, increasing numbers of leaching cycles
process relationships more complex. Carter and Davidson (1998) began to remove the features which had formed. Van Vliet-Lanoë and
summarised three problems associated with using diagnostic prop- Coutard (1984) and Van Vliet-Lanoë (1985, 1998) studied frost effects
erties, in the context of pedoarchaeological interpretations: First, in soils by subjecting soil samples to freezing cycles in the laboratory.
diagnostic properties do not necessarily develop in all soils. Second, In these studies soil micromorphology was also used at each stage to
diagnostic properties can be destroyed by subsequent processes or by reveal the number of freezing cycles necessary to produce recogniz-
post burial modification. Finally, the absence of a diagnostic feature able features.
does not necessarily mean that the process did not take place. In order to refine the signatures for specific activities, soil studies
The challenge to identifying process–response relationships is at experimental and historical sites have been used (Carter and
equifinality, whereby different suites of processes can produce the Davidson, 1998; Macphail, 1998; Carter and Davidson, 2000). In this
same final form (Simonson, 1959; Goldberg and Macphail, 2006). way signatures detected through elemental analysis and micromor-
Many soils are polygenetic and have formed under a range of phology can be linked to known land management events. For
changing environmental conditions, thus it is challenging to link example, at the experimental site of Butser Ancient Farm, southern
properties to processes for a known time period. For example, Carter England, there was an opportunity to study the micromorphology of
and Davidson (1998) and Usai (2001) have critically evaluated the use experimental floor types (domestic and stable floors) alongside
of soil micromorphology to explore whether there is a soil palynology and bulk chemical analysis (Macphail et al., 2004;
morphological response that can be related to ancient arable Goldberg and Macphail, 2006). Comparing crop yield under differing
agriculture. They concluded that the presence of dusty and silty clay land management regimes was studied at Umeå ancient farm, Sweden
textural coatings around grains in thin section cannot be linked (Viklund, 1998). There has been a useful debate in the literature about
causally to ploughing activity, even at sites of known cultivation. The the general application of templates from experimental sites to other
formation of these coatings may arise as a function of favourable soil archaeological settings (Canti et al., 2006; Macphail et al., 2006).
porosity, structure and texture (Usai, 2001), which in some soils may Studies at experimental earthworks (such as Overton Down and
be enhanced by cultivation, and in other soils may derive from non- Wareham, UK) to investigate short term (c. 30 years) pedological
anthropogenic factors, such as inherited texture from the soil parent change as a result of burial, have shown that the micro-environments
material. Even if cultivation created the necessary preconditions for created, such as relative moisture contents, degrees of compression
the formation of dusty and silty clay coatings on grains, a high organic and abundance of mesofauna, all affect decomposition rates and
matter content and high turnover by earthworms may preclude their determine the nature of post burial change. Using a combination of
preservation (Carter and Davidson, 1998; Davidson and Carter, 1998; micromorphology, organic matter, carbonate and iron contents, pH,
Usai, 2001; Davidson, 2002). Dusty and silty clay coatings are similar and magnetic susceptibility these variables were compared to a
features to those resulting from experimental studies of trampling control soil to help quantify the changes. This gives a clearer sense of
and freeze–thaw processes (Goldberg and Macphail, 2006). Indeed the changes which buried soils will have undergone at archaeological
Courty et al. (1989) gave at least six possible interpretations for dusty sites. (Crowther et al., 1996). The burial medium impacted directly on
clay coatings, exemplifying the diversity of possible explanations. decomposition rates of surface organic rich horizons in a lowland
There are a variety of approaches to the problem of equifinality in podzol (spodosol), mainly due to the different moisture holding
archaeological soils. The following sections consider: the description capacity of the construction materials (Macphail et al., 2003). Such
and interpretation of thin sections (soil micromorphology) to try to work is important because archaeological earthworks may preserve
identify morphological signatures for particular soil processes; the use buried soils, which could provide clues to the climate and land-use of
of experimental pedology, either in the laboratory, to carefully former landscapes. Without understanding how these signatures
monitor individual processes and soil responses, or at experimental might be altered by the burial process, it would be impossible to
sites such as earthworks and farms; the use of historical and control decode the buried soil archive.
sites to develop reference materials; the scientific identification of
elemental, isotopic and biomarker soil signatures. 2.1.3. Historical sites
A study of the historical landscape of Papa Stour, Shetland,
2.1.1. Thin section analysis (soil micromorphology) Scotland used a quantitative image analysis approach to thin sections,
The objective of soil micromorphology is to “interpret the in order to compare the spatial responses to documented manuring
composition and arrangement of soil components in terms of management practices (Adderley et al., 2006). Entwhistle and
particular types and orders of processes” (Kemp, 1985, pp. 5). Abrahams (1997), Entwhistle et al. (1998) used ICP-MS to investigate
Micromorphological structures can provide evidence of anthropo- land use from soils at several Scottish historical sites. The use of shell
genic processes. Gebhardt (1992) used experimental techniques to try amendments on the land showed up as Ca and Sr ‘enrichment factors’
to distinguish patterns formed by specific agricultural implements. and this was supported by documentary evidence of land husbandry
126 H. Walkington / Earth-Science Reviews 103 (2010) 122–134

techniques where shells from beach material were added to the sites with thin section analysis, integrating visual and non visual methods.
to improve soil fertility. Lipid biomarkers provided further evidence of Carbon isotopes can reveal traces of previous vegetation cover (West
anthropogenic soil enrichment. et al., 2000) and crop types (Webb et al., 2004). Simpson (1997) used
micromorphology, phosphate tests and particle size analysis to study
2.1.4. Elemental analysis relict properties which were used to identify management practices
There has been intense interest in characterising soil responses to on West Mainland, Orkney. A reference collection of known materials
ancient agricultural activity. This section reviews these developments. was used and other areas of the landscape were also tested to verify
Ancient agricultural activity is often identified through the recogni- the data.
tion of anthropogenic additions to the soil. This is achieved through the Cultivation can alter soil properties in quantifiable ways (Homberg
identification of soil signatures. Phosphate has traditionally been used et al., 2005). In some areas cultivation is represented as degradation,
to detect agricultural intensity and practices (Conway, 1983; Bjelajac in others as soil improvement. In some geographical locations
et al., 1996; Farswan and Nautiyal, 1997; Beach, 1998b; Terry et al., agricultural land management accelerates existing soil processes so
2000; Holliday and Gartner, 2007) as well as coprolites (Allen et al., an understanding of the pedogeomorphological landscape is funda-
2002). Anthropic additions of phosphate can include human refuse mental to understanding the anthropogenic signature at a site
and waste, burials, livestock dung and manure as a soil amendment. (Romans and Robertson, 1983; French, 2003). The examples above
However, the phosphate signal which remains in the soil can be have shown that ancient agricultural practices are difficult to decipher
affected by post depositional weathering, the nature of the parent from the soil record. However, elemental analysis has allowed
material and the intensity of occupation (Holliday and Gartner, 2007). significant advances to be made in the identification of chemical
Recent advances in elemental analysis have increased the potential signatures deriving from amendments, such as manure and shell. The
for identifying multi element signatures, rather than relying on identification of morphological signatures is much more difficult as a
phosphate. Entwhistle and Abrahams (1997) outlined the technique result of equifinality of form i.e. different processes producing the
of ICP-MS, for the simultaneous, multi element analysis of soils and same morphological pattern. However, used in combination with
sediments, and its limitations in archaeological contexts. ICP MS other evidence, thin sections can also reveal the presence of chemical
allows rapid site investigation and can be used for inorganic elements, e.g., the presence of calcium carbonate spherulites from
compounds and organo metallics. It is quantitative and allows the animal dung (Canti, 1998). Indeed, the increased use of multiple
archaeological scientist to appraise a large suite of elements at a site. methods has revealed the limitations of using some methods in
The authors suggest that for the analysis of phosphate, problems with isolation, such as the ‘lag’ effect of pollen in recording woodland
spectral interference mean that alternative methods should be used clearance (Dark, 2005).
for this element. A further problem is the large range in the results for The approaches adopted to try to address the problem of
different elements, with some elemental occurrence levels being an equifinality range in scale from elemental (e.g., biomarkers) to
order of magnitude greater than others. The authors suggest that it is whole sites within their landscape setting. In some cases it is
best to carry out studies at one level of magnitude, for example by impossible to resolve the problem of equifinality; however the
using atomic absorption and flame emission for elements like K, Ca challenge which it poses continues to provide opportunities to
and Mg. Soils can also be classified based on their elemental refine research approaches and further our understanding of soil
composition in terms of major and trace elements using ICP-AES development.
and this technique now dominates in more recent pedoarchaeological
studies. In this method electromagnetic radiation emitted by samples 2.2. Distinguishing property–process links where preservation is poor
is used to detect trace elements in the soil and concentrations of
elements. For example, Linderholm and Lundberg (1994) detected Yaalon has suggested that to elucidate the preservation potential
traces of human impact on Swedish soils by comparing archaeological of pedogenic properties under a variety of diagenetic conditions is a
and control sites. A framework for the identification of agricultural major research objective (1986: 592). Diagenetic changes can alter
paleosols was developed on the basis of this technique, since elements and even remove soil properties, environmental information and
found in the Bronze age archaeological site of Vistad in southern artefacts. It is important for archaeologists to be able to reconstruct
Sweden were different to those which characterised the control what artifacts and ecofacts were like, before they were disturbed or
samples. degraded. This is most challenging in situations where preservation is
poor. For pedoarchaeologists it is important to understand the
2.1.5. Biomarkers processes and pathways of artifact incorporation into the soil, post
Biomarkers are ancient molecules and include those derived from burial modifications and diagenetic change.
human use of biological material and organisms. Although these show
no visible remains, the proteins and lipids (e.g. fats, oils, resins and
2.2.1. Bioturbation and site integrity
waxes) have a diagnostic fragmentation pattern when subjected to
Much recent work has focussed on the impact of bioturbation, both
gas chromatography and can be detected using an attached mass
in preserving artefacts by incorporating them into the soil, and
spectrometer (Evershed et al., 1999). Despite the fact that degraded
subsequently impacting on site integrity.
fats can be difficult to distinguish following diagenesis, it is possible to
detect their stable isotope ratios using gas chromatography, coupled “The interaction between soil organic matter, soil micro-organisms
with an isotope ratio mass spectrometer (GC-IRMS) (Evershed et al., and soil structure and properties is extraordinarily complex. It is
1999). Lipids and biomarkers have been used as effective indicators of also the environment in which archaeological materials spend the
past manuring practice (Evershed et al., 1997). For example, they have vast majority of their existence, and it can be ignored only at the
been used to detect and distinguish terrestrial and marine animal cost of the faithfulness of archaeological reconstruction to past
residues in organic soil samples (Nolin et al., 1994). Bethell et al. reality.” (Haslam, 2005, pp. 1728)
(1994) used the molecular marker coprostanol as an indicator of
human faecal material. The sterol has ubiquitous occurrence in the Biota are commonly responsible for the burial and therefore the
environment but a specific signature was quantified by relative preservation of artefacts below the soil surface (Van Nest, 2002;
abundances and ratios. A multi method approach has also become Fowler et al., 2004). In the case of Van Nest's (2002) research in the
standard in the study of early arable land management. Simpson et al. loess deposits in Western Illinois, USA, archaeological sites had no
(1998a) coupled studies of stable carbon isotopes and free soil lipids new accumulations of aeolian material since the arrival of
 H. Walkington / Earth-Science Reviews 103 (2010) 122–134 127

paleoindians; thus artefact preservation was a function of biotic human impact. After a brief discussion of geophysical methods, the
activity. This work reminds us that some biomantles preserve identification of activity areas at sites produces a second example.
stratigraphy while others do not. Biota are important in studying Amazonian Terra preta (Indian dark soils) are dark anthrosols with
the pathways of pedogenesis (Johnson, 2002; Robertson and Johnson, high organic content, even in the subsoil, which contain artefacts.
2004; Johnson et al, 2005a,b; Johnson et al., 2008). Peacock and Fant These soils have intrigued numerous researchers (for example Smith,
(2002) considered the balance between progressive (formation of soil 1980; Eden et al., 1984; Graham, 1998; Holliday, 2004) who have tried
horizons) and regressive (removal and mixing of soil horizons) to ascertain their likely origin. Whilst early theories included lake
vectors on soil development, at a site developed in sandy soils in sedimentation and deposition of volcanic ash, these theories were
Mississippi, USA. They showed how the process of bioturbation could dismissed on spatial grounds. Similar deposits have also been found
be regarded as both progressive and regressive. This reinforces forming today in the region (Smith, 1980), so the Terra preta are now
Simonson's (1978) idea that soil development processes can operate recognised as ‘archaeological soils’ from Pre-Colombian settlement
at all times but to differing extents. The soil development pathway at (Eden et al., 1984). The accumulation of domestic waste products
the site was the combined effect of horizonation and haploidisation would account for their darker colour, higher carbon and phosphorus
processes. Grave and Kealhofer (1999: 1239) suggested that in order status than adjacent paler, less fertile soils. The Terra Preta de Indio
to understand site integrity, multiple studies of this type should be have an anthropogenic epipedon which, despite intense leaching,
carried out within a single geomorphological setting. retains a dark colour and plant nutrients (Lima et al., 2002). They have
Many authors have noted the negative impact of bioturbation on been described by McCann et al. (2001) and Woods and McCann
the stratigraphic integrity of archaeological sites, such as tree throw (1999) who noted the darkest soils (Terra preta) most probably
(Macphail and Goldberg, 1990), earthworm activity (Van Nest, 2002; resulted from habitation, whereas the brownish (Terra mulata) were
Canti, 2003), crayfish burrowing (Robertson and Johnson, 2004), more likely to be formed by the application of agricultural amend-
mammal burrows (Balek, 2002). Armour-Chelu and Andrews (1994) ments such as compost and mulch. Experimental work has shown
showed experimentally how the impact of bone burial by soil that it is the addition of charcoal to the soil that significantly increases
organisms resulted in erosion of the bones as well as significant plant growth and nutrition (Lehmann et al., 2003, 2004). It has the
vertical and lateral displacements. Canti (2003) noted that although effect of reducing the leaching of nitrogen and delaying the leaching of
this can be the case, worms tend to have burrows that they calcium and magnesium which have contributed to the long lasting
permanently inhabit, reducing the homogenising effect on the soil. fertility of these soils, and indeed to their continued use today in some
Bioturbation by worms is specific to particular species, places and soil locations (Eden et al., 1984). The extent of the soils is of interest to
conditions. Termites had preserved casts of early African cultigens archaeologists as they could form the basis of estimates of pre-
where their activity was localised (Fowler et al., 2004). Davidson colonial population. Analysis of the clay mineralogy from the pottery
(2002) quantified excremental pedofeatures in thin section to show found within the soils has shown that the Amazon floodplain was the
the extent of bioturbation in samples from cool temperate uplands source of the raw material (Lima et al., 2002). Research on Terra preta
and found that unless soils were buried and sealed from subsequent soils has been recently summarised by Glaser and Woods (2004).
diagenesis the loss of features associated with cultivation occurred This example has shown that identification of anthropic soil
within decades as a result of bioturbation. However, evidence of development in the Terra preta was confirmed through a combination
manuring was more persistent. In this environment enchytraeids approaches including spatial mapping, soil description and elemental
were the responsible biota. Catt (1987) suggested that it is most analyses. It reveals the importance of combining research methods at
useful to consider the individual relict features of a soil and to try and different scales.
sequence them in time and then relate them as precisely as possible Anthropogenic processes can be expressed as a change in the rate
to particular periods of the past. This is a micro-pedostratigraphic of soil processes, rather than as an easily identifiable human artefact,
approach (Creemens et al., 1998) and requires the use of soil such as accelerated soil erosion on hillslopes (Section 2.4.1). It is
micromorphology (Kemp, 1985; Sander, 2002). problematic to detect accelerations in existing processes when there
A particular challenge in the case of archaeological soils is in is no temporal framework to the properties produced. Furthermore,
applying stratigraphic principles, when soils represent long periods of the balance between processes may cause one property to be created
time with ongoing processes, rather than discrete events at points in at the expense of another, so the overprinting of processes on the soil
time. If archaeological sites are not isolated from contemporary creates a palimpsest. Sequencing and dating these processes can be
processes, the properties will be subject to ongoing additions, extremely complex. The situation is further complicated at sites
removals, transfers and transformations. Bioturbation is just one where archaeological soils are not isolated from contemporary
example of soil mixing. Soils can be disturbed by other factors as well processes, so that overprinting is ongoing. For example, Terra preta
as bioturbation (which also uincludes florapedoturbation), such as ice soils remain susceptible to weathering and leaching processes as well
crystal growth (cryopedoturbation), shrinkage and swelling of clays as possible later cultivation, which can have an ongoing influence on
(argillipedoturbation), seismic activity (seismipedoturbation). For a their physical and chemical characteristics (Eden et al., 1984, p.138).
recent review see Schaetzl and Anderson (2005). Recently, however, Hutson et al. (2009) used P fractionation to
This section has shown that it is important to study site integrity at distinguish ancient from modern gardening activities at a Maya site in
a variety of scales. Experimental work in the laboratory can indicate Yucatan, Mexico. This technique is effective on sites which are not
the impacts of specific soil fauna, but this needs to be coupled with isolated from contemporary anthropogenic processes. Modern phos-
landscape scale work to understand how three dimensional processes phate enrichment can be differentiated from ancient amendments.
operate. Vector and balance based models are very useful to The authors stress that the method requires very careful, site by site,
conceptualise different preservation scenarios. evaluation and needs to be correlated with other lines of evidence. In
their study, experimental approaches were used, such as using
2.3. The challenge of isolating anthropogenic soil processes limewater for soaking maize. Soil signatures from these activities
were then developed.
In order to make archaeological interpretations from soil, it is Engagement with a balance based model, such as that proposed by
important to distinguish anthropogenic impacts upon soil formation Simonson (1959) can go some way towards solving the problem of
and soil development. The Terra preta soils of Amazonia serve as the the overprinting of processes. By re-examining the spatial and causal
first example of the challenge of clearly identifying anthropic succession of features it is also possible to identify how a soil may
processes of soil formation from those which have not arisen from have evolved. The modelling of processes at experimental sites and in
128 H. Walkington / Earth-Science Reviews 103 (2010) 122–134

the laboratory is also crucial in being able to identify diagnostic origin, (less mobile elements tend to be linked to soil parent material).
features. While exploring the nature of activities which took place in a Hunter–
Gatherer cave in the USA, Homsey and Capo (2006) used this method
2.3.1. Geophysical approaches to distinguish cultural features. Micromorphology was then employed
The identification of activity areas has developed rapidly in recent to subdivide these feature types and to examine how taphonomic
years, particularly in distinguishing anthropogenic from background processes had operated on the cave sediment. Knudson et al. (2004)
signatures, where little surface evidence exists. The geophysical used ICP-AES to create a contemporary chemical signature for fish
properties of soil have been used to identify activity areas at some camps in the Arctic, with the aim of using the data as a template for
archaeological sites in a non-invasive way. This is achieved through studying archaeological sites. Sampiettro and Vattuone (2005) were
techniques such as ground penetrating radar (GPR), electrical able to differentiate burials, plant storage and animal processing as
resistance and magnetometry. With GPR users study the sub surface the major domestic activities in a formative household in Argentina by
reflectance of radio waves detected by antennae to interpret different using soil chemistry, coupled with ceramic and bone distributions.
sediments, floors and prepared surfaces (Goldberg and Macphail, The use of multi-proxy indicators is essential for such identifications.
2006). Electrical resistance techniques operate by passing an electrical In most cases micromorphology is integrated, such as combining
current through electrodes in the soil. Differences in electrical phytolith analysis with soil chemistry (Sullivan and Kealhofer, 2004),
resistance can be detected. Resistivity surveys can detect boundaries micromorphology with palynology (Tipping et al., 1994) field survey
between materials and their depths. Perturbations to the earth's and micromorphology and image analysis (Bryant and Davidson,
magnetic field over a ground surface can be detected by magnetom- 1996; Simpson et al., 1998b). Even in areas completely lacking in
etry and buried archaeological materials, particularly those relating to visible structures, such as abandoned Maasai settlements and ancient
heating activity, can contribute to this pattern e.g. pits, ditches, Tell sites, it has been possible to use combined techniques such as
hearths and other burned areas (Gaffney and Gater, 2003; Goldberg micromorphology, mineralogy and phytolith analysis, to identify
and Macphail, 2006). Although the above methods are used at activity areas such as livestock enclosures and domestic areas
archaeological sites, the approach related most specifically to soils in (Shahack-Gross et al., 2003, 2005).
archaeological contexts is magnetic susceptibility which can be The study of indoor use of space has focussed on preserved floors
measured in the field and laboratory. from archaeological sites. Residues in soil and plaster floors from
All substances possess magnetic properties (Smith, 1999). When a Mayan sites have been identified using ICP-AES (e.g. Middleton, 2004;
magnetic field is applied to soil, the degree of magnetisation is Hutson and Terry, 2006; Hutson et al., 2007). As well as phosphorus,
recorded as magnetic susceptibility. Anthropogenic inputs to soil such researchers can use the ICP-MS to identify a wide range of elements
as burnt material, enhance the signal greatly e.g. in Dark Earths such as trace and ultra trace elements. These elements have provided
(Goldberg and Macphail, 2006: 350–351). In trying to showcase the evidence of industrial and ritual use, for example at the Maya city of
variety of benefits of using magnetic susceptibility, Dalan and Cancuén, Guatemala. Cook et al. (2006) studied late classic period soil
Banerjee (1998) have shown how the technique can be useful in floors and found rare earths, gold and mercury. Paleosols were used as
delimiting areas of former occupation, understanding erosion, controls to establish background levels of elements in the environ-
identifying the application of manure, identification of buried soils ment at the time. However, soils were not found to be adequate
and interpreting past weathering and climatic regimes. They do stress controls for studies of plaster floors (Hutson and Terry, 2006). In their
the importance of using the technique in association with other study of Chunchumil in Mexico they also found that activities could be
methods. linked to levels of P and trace metal concentrations (ibid). Hutson et
Common enhancements to the magnetic susceptibility of soil al. (2007) studied outdoor space using a combination of systematic
minerals arise from topsoil production of maghaemite by biological subsurface soil sampling and ethnoarchaeological models to recon-
activity, burning and the general pedogenetic production of magnetite struct economic activities of the ancient Maya.
and maghaemite (Goldberg and Macphail, 2006). Magnetic suscep- Soil micromorphology was the basis of distinguishing anthropo-
tibility measurements can help to identify features with enhanced genic sediments used in settlement construction at Skara Brae,
magnetic properties, such as hearths and pits (Dalan and Banerjee, Orkney. It was used to identify soil processes and post depositional
1998; Sarris et al., 2004). Archaeological soils can retain very fine disturbances (Simpson et al., 2006). Researchers have analysed soil
grained superparamagnetic particles of characteristic mass and size thin sections for the intensity of activities and changes from one
produced by burning (Peters and Thompson, 1999). The detection of activity to another e.g. from fishing to arable agriculture (Simpson et
heated soils, indicative of the early use of fire, (Morinaga et al., 1999; al., 2005) or changes in manuring practices and their intensity
Canti and Linford, 2001a,b) is a further application. Using soil (Guttman et al., 2006). Creemens (2005) used soil micromorphology
micromorphology, coupled with an experimental approach, it is to distinguish anthropogenic from other taphonomic processes when
possible to describe different types of burnt layer, and the effects of studying Cotiga mound, W. Virginia, USA. Distinguishing archaeolog-
different temperatures and conditions can be analysed (Courty et al., ical layers at sites from a sediment sequence is a strength of the
1989: 107). methodology. Exaltus and Miedema (1994) were able to distinguish
Pedogenic iron content has been used to distinguish paleosols in nine types of archaeological layer from Neolithic coastal sites in the
Chinese loess sequences (for example, Tang et al., 2003). However, Netherlands and to turn their characteristics into interpretation
Evans and Heller (2001) warn of making assumptions about paleosol criteria, an approach which could not have been possible without
magnetic properties. Working in Siberia they found that wind vigour micromorphology.
can dominate over pedogenesis such that a paleosol can have a lower However, micromorphological interpretation has its limitations
magnetic signals than overlying aeolian deposits. Likewise, in some (Canti et al., 2006), There is difficulty linking micromorphological
waterlogged sediments they discovered that gleying was associated features directly to human activities (Arpin et al., 1998). Stratigraphic
with a loss of magnetic minerals. principles at the scale of individual pores in the soil may be necessary
to fully resolve such issues (Kemp, 1985; Sander, 2002; French, 2003)
2.3.2. Activity areas and a two dimensional thin section will not always contain the
The recent growth of elemental analysis has enabled multiple soil evidence required to validate an interpretation (Stoops, 2003: 11).
samples to be taken, representing large surfaces. This has helped to Image analysis uses thin sections to try to quantify soil micromor-
identify activity areas and thus make inferences about past human use phology. It has been used in soil science to rapidly quantify visible
of space. ICP-AES can measure mobile elements of anthropogenic patterns, such as pore space (Lima et al., 2006; Marcelino et al., 2006).
 H. Walkington / Earth-Science Reviews 103 (2010) 122–134 129

Although it can enhance description and interpretation in archaeo- In northwest Europe, soil erosion accompanied early agriculture,
logical applications, the approach requires careful interpretation, and and has also been studied through multi-method approaches. For
should be used in conjunction with micromorphology and other example, in Sweden Lagerås and Sandgren (1994) combined
standard tests (Bryant and Davidson, 1996; Matthews et al., 1997; palynology and magnetic studies to show how the changes in
Goldberg and Macphail, 2006). Francus (2004) provides a recent vegetation as a result of the introduction of agriculture, corresponded
review. to episodes of erosion, as highlighted by changes in the magnetic
This section has shown that the nature of the anthropogenic properties of the peat profiles. Attempts have been made to identify
process being investigated impacts the effectiveness of methods to human induced erosional events using multiproxy indicators (Van-
unambiguously identify it. The presence of trace metals can reveal niere et al., 2003). In Iceland, Simpson et al. (2004) found that winter
significant findings about human activities in the past, but do little to grazing was responsible for past landscape degradation, but that this
inform our understanding of soil processes. In situations where soils was mediated by inherent soil susceptibility in some geographical
have been subjected to ongoing processes a combination of methods areas. The anthropogenic process manifested itself as an acceleration
is required to resolve pedoarchaeological questions. of an existing pedogeomorphological process.
In the eastern Mediterranean and Levant, the impact of humans on
2.4. Difficulties of soil interpretation due to the interdependency of the accelerated soil erosion has required landscape scale analysis of
soil forming factors paleosols and sediment accumulation. Regional variation, in the
timing and amount of erosion and aggradation, has characterised
The soil forming factors are interdependent, therefore it is difficult these studies (French and Whitelaw, 1999; Wilkinson, 1999; Butzer,
to separate the effects of one from the other when interpreting 2005; Wilkinson, 2005; Beach and Luzzadder-Beach, 2008). Paleosol
ancient soils. To address this problem, the choice of field location for stratigraphy and burial events have helped to more securely date the
soil studies is important. Maher et al. (2002a,b, 2003) used the vast events which have occurred in different valley and mountain areas. In
area of the Russian Steppe to isolate climate as a variable and to Western Jordan, Cordova (2008) used relative dating of diagnostic
document how it produced a magnetic signal that could be employed lithics and ceramics, contained in paleosols, to contribute to an
as a palaeorainfall proxy in the soils. Nevertheless, the likelihood of understanding of stratigraphy in exposures. Wadi paleosols provided
finding archaeological soils in areas which have experienced only one evidence of stable periods. Paleosols were also used by Beach and
major soil forming regime or interval is limited. Luzzadder-Beach (2008) to ascertain depths of aggradation. From
The effect of time in causing soils to ‘mature’ can often be difficult sections in Turkey, they noted that aggradation in the Hellenistic to
to separate from the impacts of a particular climatic regime. Where Roman periods, was more than twice as severe as anything
time can be isolated as the only variable across a landscape, it is the experienced previously, or since. Cordova et al.'s research on plateau
basis of a chronosequence (Harden, 1982). However, in similar areas (2005) and wadis (2008), suggests that rapid agricultural
environments, different soil forming intervals can produce different development may have altered slope stability, and resulted in an
features. This is further complicated by the tendency to assume that increase of colluvial materials on the floodplain, due to a combination
ancient soils were in equilibrium with their environment, despite the of land use mismanagement and precipitation variability. Several
fact that soil forming environments have undergone very rapid authors agree that soil erosion resulted from a combination of climatic
change during the Quaternary (Lowe and Walker, 2006). In areas variability and susceptibility created by changing land use (Butzer,
where these changes created sedimentation, such as areas of loess 2005; Wilkinson, 2005; Casana, 2008; Cordova, 2008). For example,
accumulation, the balance between the processes of soil formation Casana (2008), working in southern Turkey, noted a lag of over
and sedimentation also need to be considered (Kemp, 2001). In 400 years between the onset of upland agricultural production (in the
aggrading landscapes, soil studies can help to define the stratigraphy Hellenistic, Roman and late Roman period) and aggradation. He
(Wagner and McAvoy, 2004). Artefacts can even be used to date the suggests that, coupled with the new land use conditions required for
sediments that contain them, which in turn can be used to ascertain a the production of olives, cereals and vines, extreme levels of
chronological framework for palaeoenvironmental change (Ripley, precipitation could have caused the aggradation. Environmental
1998). stress, resulting from regional aridity and increased population
This section now reviews the problem of using pedoarchaeological density in Southern Turkey, led to increased rates of erosion and
evidence to try to separate the anthropogenic effects of land sedimentation. Unambiguously identifying anthropogenic impacts
management and changing landuse, from climatic variability, at a upon the landscape from other factors such as climatic variability
range of sites. remain problematic. However, a landscape scale pedogeomorpholo-
gical approach (Conacher and Dalrymple, 1977; Gerrard, 1992),
2.4.1. Accelerated soil erosion linked to multiproxy indicators for dating and climate reconstruction,
In Meso America, Mayan agriculture and its impacts upon erosion can help build a detailed picture of past processes.
and landscape aggradation events has been a focus of study (Beach,
1998a; Beach et al., 2006; Johnson et al., 2007; Beach et al., 2008). 2.5. Difficulties with soil dating
Distinct epochs of accelerated soil erosion have been identified in the
central Maya lowlands. Pioneer farmers have created the greatest The isolation of time dependent soil properties, for example,
levels of erosion, with lower levels than expected being associated degree of weathering (Anderton, 1999; Schuldenrein et al., 2004;
with the highest density of population in the area (Beach et al., 2006). Stafford, 2004), and clay accumulation (Chartres, 1980; van Andel,
It has been possible to use soil to identify the areas being used for food 1998; Frolking and Lepper, 2001; Schuldenrein et al., 2004) has
production, with stable carbon isotope analysis of soil organic matter, enabled the creation of relative dating sequences of soils. Paleosols
providing evidence of agricultural activity. In some places, aggrading can also provide a stratigraphic marker horizon and thus a relative
sediments have become trapped in depressions in the landscape and dating framework (e.g. Chittleborough, 1991; Bronger and Heinkele,
preserved paleosols (Beach et al., 2008). Micromorphology has been 1995; Butzer, 2004). Where paleosols do not have sufficient spatial
used to explore variations in land use intensity. At the Ecuadorian extent to provide a stratigraphic marker, their properties can
raised field system of Hacienda Zuleta, Wilson et al. (2002) compared sometimes still be used to create a chronosequence.
the land management practices of Hispanic and pre-Hispanic Soil radiocarbon dating is a technique which has severe limitations
societies. Relict features at a micro scale showed that land use in archaeological contexts due to a range of factors: the dynamic role
intensity was greater in pre Hispanic society. which soil plays in the global carbon cycle; the dangers of post-burial
130 H. Walkington / Earth-Science Reviews 103 (2010) 122–134

contamination; and the long soil forming intervals which may be tions. It has been speculated that in the future the carbon isotopes in
involved are three examples. Furthermore, the significant period it the calcium carbonate of earthworm granules may yield an absolute
takes for soil to form does not lend itself to yielding a single date. date of deposition (Canti, pers. comm.), although the difficulty may be
Radiocarbon dates for soil organic matter show great spatial identifying the origin of the carbon from the mix of potential sources
variability, particularly with depth (Matthews, 1980; Matthews and available in the soil environment (Hedges, 1994). There are two
Dresser, 1983; Matthews, 1985). Schuldenrein et al. (2004) and Botha further alternatives: the use of natural remanent magnetism dating
and Fedoroff (1995) have used soil carbonates and humic and fulvic (using bacterial magnetite to date a change in environmental
acids for the radiocarbon dating of paleosol organic matter. The conditions (Linford et al., 2005); and optically stimulated lumines-
composition of paleosols can change significantly over time; however, cence (OSL) dating in plaggen agriculture (Bokhorst et al., 2005) and
in some cases, the chemical structure and composition of humic and agricultural areas with rapid Holocene loess accumulation rates
fulvic acids held within them are preserved sufficiently for dating (Roberts et al., 2001) to date the burial event.
(Calderoni and Schnitzer, 1984). The molecular complexity of Dating is problematic where there has been movement of
paleosol humic acids increases with the length of burial as a result materials between horizons. At some sites, the use of independent
of oxidative degradation (Schnitzer and Calderoni, 1985). Favier- dating techniques has been applied in order to check existing dates.
Dubois (2003) recently used the oxidisable carbon ratio (OCR), which For example, Feathers et al. (2006) used OSL to check radiocarbon
gives a mean residence time of organic matter in soil horizons, to date dates at Cactus Hill, Virginia, as a way of determining if the deposits
a macro-regional scale chronostratigraphic soil in Patagonia. This soil were mixed. The evidence suggested small scale turbation and slight
was used to evaluate long term changes in archaeological distribu- mixing of sandy material. At other sandy sites, susceptibility to

Fig. 1. A generalised interpretive framework for pedoarchaeology. A summary of the stages involved in interpreting soil data from archaeological sites, drawn from a review of best
practice approaches from the current literature.
 H. Walkington / Earth-Science Reviews 103 (2010) 122–134 131

alteration of the position of buried artefacts has been noted evidence of overprinting of processes? Which other soil properties
(Wandsneider, 1988; Mayer, 2002; Peterson and Mohler, 2002; could have formed the properties found? Can anthropogenic soil
Johnson et al., 2008). forming processes and/or impacts be unambiguously identified? If
artefacts are found at the site, the way in which they were
2.6. Summary of challenges incorporated should also be established, as this can relate to pedogenic
processes. Recognition of the balance between these processes may
This paper has considered five major challenges which limit the provide clues to the environmental conditions from which an initial
interpretations that can be made from archaeological soils. These interpretation could be made (Stage 3). At this interpretive stage it
challenges are strongly interrelated. may be necessary to collect more data, or return to Stage 1 in order to
As this review has revealed, recent developments in pedoarch- study some soil properties in more detail. Once we can link soil
aeology go some way to overcoming the five pedological challenges properties to soil processes it is possible to make an assessment which
outlined in this paper. The use of experimental pedology to isolate leads to interpretation. Interpretations can address the pathway of soil
processes in the laboratory, whilst still maintaining the 3D integrity of development at the site, the palaeoenvironmental changes in
the soil has proved to be essential in developing our understanding of evidence, and the nature of the human use of the site. Interpretations
soil property process relationships. The use of experimental sites and can then be tested against the initial description, other data from the
historical sites, with documentary evidence of land management site, and from other published sites and experimental data (Stage 4).
practices, analogue and control sites and soils, all allow anthropogenic The use of an interpretive model, which recognises that soil
processes to be related to real soil responses. The identification of properties arise from the prevailing balance between soil processes, is
isotopes and biomarkers, to provide evidence for anthropic additions, central to pedoarchaeological work. The principles behind it revolve
works very well at a smaller scale, in the laboratory. At a landscape around ascertaining the responsible soil processes at each stage. They
scale, the identification of buried soils has allowed studies of erosion also imply an iterative approach, connecting the landscape scale
events and has revealed that it is necessary to engage with mapping of pedogeomorphological context of the site, with small scale, some-
paleosols, and inspection of transported soil materials, to reconstruct times microscopic, scientific techniques applied to soils at carefully
paleopedogeomorphological processes. In order to date soil more chosen locations.
securely a combination of relative and absolute dating techniques can
be applied to discrete soil events, such as burial. 4. Conclusion
The spatial nature of many of these techniques is fundamental to
understanding the relationship between soil processes and the This paper has discussed a selection of new developments in soil
properties that they create. The mere presence or absence of chemical research within archaeological contexts that have shed light on many
elements, though it reveals much information in terms of indoor persistent challenges to pedoarchaeological interpretation. Develop-
activity spaces, is more limited in the landscape because of the three ments in elemental analysis, in particular, have shown significant
dimensional nature of processes in the natural environment. advances in delimiting and understanding activity areas and human
Therefore a procedure for linking soil properties to soil processes use of space in the past. Micromorphology also remains a valuable
through a spatial framework is provided next. technique, used in combination with other methods, as it retains the
spatial integrity of soils and archaeological deposits. The literature
3. A framework for interpreting soils at archaeological sites recognises that a mixed method approach is the only way to generate
robust findings. The paper has outlined a conceptual framework
The importance of returning to the soil processes at each stage of which acknowledges challenges in pedoarchaeological interpretation
pedoarchaeological interpretation cannot be understated. It is crucial and shows some of the current approaches to addressing them.
in developing robust findings from soils at archaeological sites. Fig. 1 is Making this interpretive framework explicit is important, because it
an interpretive framework, which outlines the current best practice guides the user away from a simplistic notion that interpretations can
approach in pedoarchaeological work. Without a clear framework for result directly from a description of soil properties at archaeological
interpretation researchers can be tempted to rush to interpretations sites. Instead, soil processes are the key to understanding, and it is
from an initial description of soil properties. Incorrect interpretations essential to engage with the way that they can interact with each
can arise where equifinality and property process relationships are other to produce a range of properties. Soil processes, whether
ignored. The successful application of new approaches to the study of relating to pedogenesis, diagenetic change or anthropogenic inputs,
archaeological soils requires an explicit interpretive framework which removals, transfers or transformations, are the foundation for an
stresses the importance of linking soil properties with the processes accurate interpretation.
which formed them. Fig. 1 shows how the foundation of an accurate Pedology is an important component of geoarchaeological work.
interpretation is a rigorous description. Soil descriptions cannot be Greater understanding of the taphonomic processes that have
related directly to interpretations of environmental change or human operated at a site, both before and since artefacts were deposited,
impact, even where these are suggested by other data at an requires an understanding of pedogeomorphology and paleopedol-
archaeological site, unless the link between soil properties and the ogy. In trying to resolve pedoarchaeological problems, the use of
processes that have formed them has been identified. The plethora of techniques which increase our understanding of the three dimen-
new techniques available for the description of soils at more detailed sional nature of soil processes and their resultant properties is
levels cannot replace an understanding of the underlying soil essential. These approaches must be understood in a pedogeomor-
processes which formed them. phological context, at a landscape scale. Experimental work, whether
Fig. 1 shows a generalized approach to the interpretive process in the field or laboratory, can only increase our understanding of soil
which researchers in pedoarchaeology strive to adopt. It begins with processes, but a further challenge lies in being able to steer work
recognition and description of soil properties and other materials towards a greater understanding of processes in three dimensions,
found within the soil and the links between them. The soil properties rather than two.
are described in Stage 1. They are then explained by reference to the
processes which have formed them at Stage 2. Evidence for these References
processes should be found. Next the balance between these processes
Adderley, W.P., Simpson, I.A., Davidson, D.A., 2006. Historic landscape management: a
is explored. It is at this stage that questions relating to the five validation of quantitative soil thin-section analyses. Journal of Archaeological
challenges outlined in this paper can be posed. For example – is there Science 33, 320–334.
132 H. Walkington / Earth-Science Reviews 103 (2010) 122–134

Akamigbo, F.O.R., Dalrymple, J.B., 1985. Experimental simulation of the results of clay Chittleborough, D.J., 1991. Indices of weathering for soils and paleosols formed on
translocation in the B horizons of soils; the formation of intrapedal cutans. Journal silicate rocks. Australian Journal of Earth Sciences 38, 115–120.
of Soil Science 36, 401–409. Conacher, A.J., Dalrymple, J.B., 1977. The nine unit landsurface model: an approach to
Allen, S.D.M., Almond, M.J., Bell, M.G., Hollins, P., Marks, S., Mortimore, J.L., 2002. pedogeomorphic research. Geoderma 18, 1–154.
Infrared spectroscopy of the mineralogy of coprolites from Brean Down: evidence Conway, J.S., 1983. An investigation of soil phosphorus distribution within occupation
of past human activities and animal husbandry. Spectrochim. Acta Part A 58, deposits from a Romano-British hut group. Journal of Archaeological Science 10,
959–965. 991–1008.
Anderton, J.B., 1999. The soil artefact context model: a geoarchaeological approach to Cook, D.E., Kovacevich, B., Beach, T., Bishop, R., 2006. Deciphering the inorganic chemical
palaeoshoreline site dating in the upper peninsula of Michigan, USA. Geoarchaeol- record of ancient human activity using ICP-MS: a reconnaissance study of late Classic
ogy: An International Journal 14 (3), 265–288. soil floors at Cancuén, Guatemala. Journal of Archaeological Science 33, 628–640.
Armour-Chelu, M., Andrews, P., 1994. Some effects of bioturbation by earthworms Cordova, C.E., 2008. Floodplain degradation and settlement history in Wadi al-Wala and
(Oligochaeta) on archaeological sites. Journal of Archaeological Science 21 (4), Wadi ash-Shallalah, Jordan. Geomorphology 101 (3), 443–457.
433–443. Cordova, C., Foley, C., Nowell, A., Bisson, M., 2005. Landforms, sediments, soil
Arpin, T., Macphail, R.I., Boscian, G., 1998. Conference Report: Summary of the Spring development, and prehistoric site settings on the Madaba-Dhiban Plateau, Jordan.
1998 Meeting of the Working Group on Archaeological Soil Micromorphology, Geoarchaeology 20 (1), 29–56.
February 27–March 1, 1998. Geoarchaeology 13 (6), 645–647. Courty, M.A., Goldberg, P., Macphail, R.I., 1989. Soils and micromorphology in
Balek, C.L., 2002. Buried artefacts in stable upland sites and the role of bioturbation: a archaeology. Cambridge Manuals in Archaeology. Cambridge University Press,
review. Geoarchaeology: An International Journal 17 (1), 41–51. Cambridge.
Beach, T., 1998a. Soil catenas, tropical deforestation, and ancient and contemporary soil Creemens, D.L., 2005. Micromorphology of Cotiga Mound, West Virginia. Geoarchaeol-
erosion in the Petén, Guatemala. Physical Geography 19 (5), 378–405. ogy: An International Journal 20 (6), 581–597.
Beach, T., 1998b. Soil constraints on Northwest Yukatan, Mexico: pedoarchaeology and Creemens, D.L., Hart, J.P., Darmody, R.G., 1998. Complex Pedostratigraphy of a Terrace
Maya subsistence at Chunchucmil. Geoarchaeology: An International Journal 13 Fragipan at the Memorial Park Site, Central Pennsylvania. Geoarchaeology: An
(8), 759–791. International Journal 13 (4), 339–359.
Beach, T.P., Luzzadder-Beach, S., 2008. Geoarchaeology and aggradation around Kinet Crowther, J., Macphail, R.I., Cruise, G.M., 1996. Short-term post burial change in a humic
Höyük, an archaeological mound in the eastern Mediterranean, Turkey. Geomor- Rendzina Soil, overton down experimental earthwork, Wilshire, England. Geoarch-
phology 101, 416–428. aeology: An International Journal 11 (2), 95–117.
Beach, T., Dunning, N., Luzzadder-Beach, S., Cook, D.E., Lohse, J., 2006. Impacts of the Cruise, G.M., Macphail, R.I., 2000. Microstratigraphical signatures of experimental rural
ancient Maya on soils and soi erosion in the central Maya Lowlands. Catena 65, occupation deposits and archaeological sites. In: Roskams, S. (Ed.), Interpreting
166–178. Stratigraphy. Archaeopress, Oxford, pp. 183–191.
Beach, T., Luzzadder-Beach, S., Dunning, N., Cook, D., 2008. Human and natural impacts Dalan, R.A., Banerjee, S.K., 1998. Solving archaeological problems using techniques of
on fluvial and karst depressions of the Maya Lowlands. Geomorphology 101, soil magnetism. Geoarchaeology: An International Journal 13 (1), 3–36.
308–331. Dalrymple, J.B., Theocharopoulos, S.P., 1987. Intrapedal cutans – lateral differences in
Bethell, P.H., Goad, L.J., Evershed, R.P., Ottaway, J., 1994. The study of molecular markers their properties and their spatial clustering. Geoderma 41, 149–180.
of human activity: the use of coprostanol in the soil as an indicator of human faecal Dark, P., 2005. Mid- to late-Holocene vegetational and land-use change in the Hadrian's
material. Journal of Archaeological Science 21 (5), 619–632. Wall region: a radiocarbon-dated pollen sequence from Crag Lough. Northumber-
Birkeland, P.W., 1984. Soils and Geomorphology. Oxford University Press, New York. land, England Journal of Archaeological Science 32, 601–618.
Bjelajac, V., Luby, E., Ray, R., 1996. A validation test of a field-based phosphate analysis Davidson, D.A., 2002. Bioturbation in old arable soils: quantitative evidence from soil
technique. Journal of Archaeological Science 23 (2), 243–248. micromorphology. Journal of Archaeological Science 29, 1247–1253.
Bokhorst, M.P., Duller, G.A.T., Van Mourik, J.M., 2005. Optical dating of a Fimic Anthrosol Davidson, D.A., Carter, S.P., 1998. Micromorphological evidence of past agricultural
in the southern Netherlands. Journal of Archaeological Science 32, 547–553. practices in cultivated soils: the impact of a traditional agricultural system on soils
Botha, G.A., Fedoroff, N., 1995. Paleosols in late Quaternary colluvium, northern in Papa Stour, Shetland. Journal of Archaeological Science 25 (9), 827–838.
KwaZulu-Natal. South Africa Journal of African Earth Sciences 21 (2), 291–311. Eden, M.J., Bray, W., Herrera, L., McEwan, C., 1984. Terra preta soils and their
Journal of Archaeological Science 25 (11),:1103-1110. archaeological context in the Caqueta basin of South East Colombia. American
Bronger, A., Heinkele, T., 1995. Loess-paleosol sequences as witnesses of palaeoclimatic Antiquity 49 (1), 125–140.
records? The Kashmir valley as an example. Memoirs- Geological Society of India Entwhistle, J.A., Abrahams, P.W., 1997. Multi-element analysis of soils and sediments from
32, 109–123. Scottish historical sites. The potential of Inductively Coupled Plasma Mass Spectrom-
Brooks, S.M., 2003. Slopes and slope processes: research over the past decade. Progress etry for rapid site investigation. Journal of Archaeological Science 24 (5), 407–416.
in Physical Geography 27 (1), 130–141. Entwhistle, J.A., Abrahams, P.W., Dodgshon, R.A., 1998. Multi-element analysis of soils
Bryant, R.G., Davidson, D.A., 1996. The use of image analysis of old cultivated soils: an from Scottish historical sites. interpreting landuse history through the physical and
evaluation based on soils from the Island of Papa Stour, Shetland. Journal of geochemical analysis of soil. Journal of Archaeological Science 25 (1), 53–68.
Archaeological Science 23, 811–822. Evans, M.E., Heller, F., 2001. Magnetism of loess/paleosol sequences: recent develop-
Butzer, K.W., 2004. Coastal eolian sands, paleosols, and Pleistocene geoarchaeology of ments. Earth Science Reviews 54 (1–3), 129–144.
the Southwestern Cape, South Africa. Journal of Archaeological Science 31, Evershed, R.P., Bethell, P.H., Walsh, N.J., 1997. 5β-Stigmastanol and related 5β-stanols as
1743–1781. biomarkers of manuring: analysis of modern experimental material and assessment
Butzer, K., 2005. Environmental history in the Mediterranean world: cross-disciplinary of the archaeological potential. Journal of Archaeological Science 24, 485–495.
investigation of cause-and-effect for degradation and soil erosion. Journal of Evershed, R.P., Dudd, S.N., Charters, S., Mottram, H., Stott, A.W., Raven, A., van Bergen,
Archaeological Science 32 (12), 1773–1800. P.F., Bland, H.A., 1999. Lipids as carriers of anthropic signals from prehistory.
Calderoni, G., Schnitzer, M., 1984. Effects of age on the chemical structure of paleosol Philosophical Transactions of the Royal Society of London. Series B: Biological
humic acids and fulvic acids. Geochimica et Cosmochimica Acta 48, 2045–2051. Sciences 354, 19–31.
Canti, M.G., 1998. The micromorphological identification of fgeacal spherulites from Exaltus, R.P., Miedema, R., 1994. A micromorphological study of four Neolithic sites in
archaeological and modern materials. Journal of Archaeological Science 25 (5), the Dutch coastal provinces. Journal of Archaeological Science 21 (3), 289–301.
435–444. Farswan, Y.S., Nautiyal, V., 1997. Investigation of Phosphorous enrichment in the burial
Canti, M.G., 2003. Earthworm activity and archaeological stratigraphy: a review of soil of Kumaun, Mid Central Himalaya, India. Journal of Archaeological Science 24
products and processes. Journal of Archaeological Science 30, 135–148. (3), 251–258.
Canti, M.G., Linford, N., 2001a. The effects of fire on archaeological soils and sediments: Favier-Dubois, C.M., 2003. Late Holocene climatic fluctuations and soil genesis in
temperature and colour relationships. Proceedings of the Prehistoric Society 66, southern Patagonia: effects on the archaeological record. Journal of Archaeological
385–395. Science 30, 1657–1664.
Canti, M.G., Linford, N., 2001b. Geophysical evidence for fires in antiquity: preliminary Feathers, J.K., Rhodes, E.J., Hot, S., Mcavoy, J.M., 2006. Luminescence dating of sand
results from an experimental study. Paper given at the EGS XXIV General Assembly deposits related to the late Pleistocene human occupation at the Cactus Hill Site,
in The Hague, April 1999. Archaeological Prospection 8, 211–225. Virginia, USA. Quaternary Geochronology 1, 167–187.
Canti, M., Carter, S., Davidson, D., Limbrey, S., 2006. Problems of unscientific method Fowler, K.D., Greenfield, H.J., Van Schalkwyk, L.O., 2004. The effects of burrowing
and approach in “Archaeological soil and pollen analysis of experimental floor activity on archaeological sites: Ndondondwane, South Africa. Geoarchaeology: An
deposits; with special reference to Butser Ancient farm, Hampshire, UK” by R.I. International Journal 19 (5), 441–470.
Macphail, G.M. Cruise, M. Allen, J. Linderholm, P. Reynolds. Journal of Archaeolog- Francus, P., 2004. Image analysis, sediments and paleoenvironments. Developments in
ical Science 33, 295–298. palaeoenvironmental research, vol. 7. Kluwer Academic Publisher, Dordrecht;
Carter, S.P., Davidson, D.A., 1998. An evaluation of the contribution of soil Boston. 300 pp.
micromorphology to the study of ancient arable agriculture. Geoarchaeology: An French, C., 2003. Geoarchaeology in action: studies in soil micromophology and
International Journal 13 (6), 535–547. landscape evolution. Routledge, Abingdon. 291pp.
Carter, S.P., Davidson, D.A., 2000. A Reply to Macphail's comments on “An Evaluation of French, C.A.I., Whitelaw, T.M., 1999. Soil erosion, agricultural terracing and site
the Contribution of Soil Micromorphology to the Study of Ancient Arable formation processes at Markiani, Amorgos, Greece: the micromorphological
Agriculture. Geoarchaeology: An International Journal 15 (5), 499–502. perspective. Geoarchaeology: An International Journal 14 (2), 151–189.
Casana, J., 2008. Mediterranean valleys revisited: Linking soil erosion, land use and Frolking, T.A., Lepper, B.T., 2001. Geomorphic and pedogenic evidence for bioturbation
climate variability in the Northern Levant. Geomorphology 101 (3), 429–442. of artefacts at a multicomponent site in Licking County, Ohio, USA. Geoarchaeology:
Catt, J.A., 1987. Paleosols. Progress in Physical Geography 11 (14), 487–510. An International Journal 16 (3), 243–262.
Chartres, C.J., 1980. A quaternary soil sequence in the Kennet Valley, central southern Gaffney, C., Gater, J., 2003. Revealing the buried past, geophysics for archaeologists.
England. Geoderma 23, 125–146. Tempus, Stroud, Gloucestershire.
 H. Walkington / Earth-Science Reviews 103 (2010) 122–134 133

Gebhardt, A., 1992. Micromorphological analysis of soil structural modification caused Lehmann, J., PereiradaSilva Jr., J., Steiner, C., Nehls, T., Zech, W., Glaser, B., 2003. Nutrient
by different cultivation implements. In: Anderson, P.C. (Ed.), Prehistoire de availability and leaching in an archaeological Anthrosol and a Ferralsol of the
L'Agriculture: nouvelles approaches experimentales et ethnographiques. Centre central Amazon basin: fertility, manure and charcoal amendments. Plant and Soil
Nationale de la Recherche Scientifique, Paris, pp. 373–392. 249, 343–357.
Gerrard, J., 1992. Soil geomorphology: an integration of pedology and geomorphology. Lehmann, J., Kern, D.C., Glaser, B., Woods, W.I. (Eds.), 2004. Amazonian Dark Earths:
Chapman Hall, London. 292pp. Origins, Properties, Management. Kluwer Academic Publishers, New York.
Glaser, B., Woods, W.I. (Eds.), 2004. Amazonian Dark Earths: Exploration in Space and Leigh, D.S., 1998. Evaluating artifact burial by eolian versus bioturbation processes,
Time. Springer, New York. South Carolina Sandhills, USA. Geoarchaeology: An International Journal 13 (3),
Goldberg, P., Macphail, R.I., 2003. Short contribution: strategies and techniques in 309–330.
collecting micromorphological samples. Geoarchaeology 18 (5), 571–578. Leopold, M., Völkel, J., 2007. Colluvium: definition, differentiation, and possible suitability
Goldberg, P., Macphail, R.I., 2006. Practical and theoretical geoarchaeology. Blackwell for reconstructing Holocene climate data. Quaternary International 162–163, 133–140.
Science Ltd, Oxford. 454pp. Lima, H.N., Schaefer, C.E.R., Mello, J.W.V., Gilkes, R.J., Ker, J.C., 2002. Pedogenesis and
Graham, E., 1998. Metaphor and metamorphism: some thoughts on environmental pre-Colombian land use of “Terra Preta Anthrosols” (“Indian black earth”) of
metahistory. In: Balée, W. (Ed.), Advances in Historical Ecology. Columbia Western Amazonia. Geoderma 110 (1/2), 1–17.
University Press, New York. Lima, H.V., Silva, A.P., Santos, M.C., Cooper, M., Romero, R.E., 2006. Micromorphology
Grave, P., Kealhofer, L., 1999. Assessing bioturbation in Archaeological sediments using and image analysis of a hardsetting Ultisol (Argissolo) in the state of Ceará (Brazil).
soil morphology and phytolith analysis. Journal of Archaeological Science 26, Geoderma 132 (3–4), 416–426.
1239–1248. Linderholm, J., Lundberg, E., 1994. Chemical characterisation of various archaeological
Guttman, E.B., Simpson, I.A., Davidson, D.A., 2006. The management of arable land from soil samples using main and trace elements determined by inductively coupled
prehistory to the present: case studies from the Northern Isles of Scotland. plasma atomic emission spectrometry. Journal of Archaeological Science 21 (3),
Geoarchaeology: An International Journal 21 (1), 61–92. 303–314.
Hall, R.D., 1999. A comparison of surface soils and buried soils: factors of soil Linford, N., Linford, P., Platzman, E., 2005. Dating environmental change using magnetic
development. Soil Science 164 (4), 264–287. bacteria in archaeological soils from the upper Thames Valley, UK. Journal of
Harden, J.W., 1982. A quantitative index of soil development from field descriptions: Archaeological Science 32, 1037–1043.
examples from a chronosequence in central California. Geoderma 28, 1–28. Lowe, J.J., Walker, M.J.C., 2006. Reconstructing Quaternary Environments (Third
Haslam, M., 2005. The decomposition of starch grains in soils: implications for Edition). Pearson Education Ltd., London. 389pp.
archaeological residue analysis. Journal of Archaeological Science 31, 1715–1734. Macphail, R.I., 1987. A Review of Soil Science in Archaeology in England. In: Keeley,
Heathcote, J.L., 2002. An investigation of the Pedosedimentary Characteristics of H.C.M. (Ed.), Environmental Archaeology and Regional Review Vol II Occasional
Deposits Associated with Managed Livestock. Unpublished PhD thesis, Institute of Paper No. 1. Historic Buildings and Monuments Commission for England,
Archaeology, London. London, pp. 332–379.
Hedges, R.E.M., 1994. Radiocarbon dating of soils in archaeology. SEESOIL (Southeast Macphail, R.I., 1998. A reply to Carter and Davidson's “An Evaluation of the Contribution
England Soils Discussion Group) 10, 5–11. of Soil Micromorphology to the Study of Ancient Arable Agriculture”. Geoarchaeol-
Holliday, V.T., 2004. Soils in Archaeological Research. Oxford University Press, Oxford. ogy: An International Journal 13 (6), 549–564.
Holliday, V.T., Gartner, W.G., 2007. Methods of soil P analysis in archaeology. Journal of Macphail, R.I., Goldberg, P., 1990. The micromorphology of tree subsoil hollows: their
Archaeological Science 34, 301–333. significance to soil science and archaeology. In: Douglas, L.A. (Ed.), Soil-
Homberg, J.A., Sandor, J.A., Norton, J.B., 2005. Anthropogenic influences on Zuni Micromorphology: A Basic and Applied Science Developments in Soil Science.
agricultural soils. Geoarchaeology: An International Journal 20 (7), 661–693. Elsevier, Amsterdam, pp. 425–429.
Homsey, L.K., Capo, R.C., 2006. Integrating geochemistry and Micromorphology to Macphail, R.I., Crowther, J., Acott, T.G., Bell, M.G., Cruise, G.M., 2003. The experimental
Interpret Feature Use at Dust Cave, a Paleoindian Through Middle-Archaic Site earthwork at Wareham, Dorset after 33 years: changes to the buried LFH and Ah
in Northwest Alabama. Geoarchaeology: An International Journal 21 (3), horizons. Journal of Archaeological Science 30 (5), 77–93.
237–269. Macphail, R.I., Cruise, G.M., Allen, M.J., Linderholm, J., Reynolds, P., 2004. Archaeological
Hutson, S.R., Terry, R.E., 2006. Recovering social and cultural dynamics from plaster soil and pollen analysis of experimental floor deposits; with special reference to
floors: chemical analyses at ancient Chunchucmil, Yucatan, Maxico. Journal of Butser Ancient farm, Hampshire, UK. Journal of Archaeological Science 31, 175–191.
Archaeological Science 33, 391–404. Macphail, R.I., Cruise, G.M., Allen, M.J., Linderholm, J., 2006. A rebuttal of the views
Hutson, S.R., Stanton, T.W., Magnoni, A., Terry, R.E., Craner, J., 2007. Betond the buildings: expressed in “Problems of unscientific method and approach in Archaeological soil
formation processes of ancient Maya houselots and methods for the study of non and pollen analysis of experimental floor deposits; with special reference to Butser
architectural space. Journal of Anthropological Archaeology 26, 442–473. Ancient farm, Hampshire, UK by R. I. Macphail, G. M. Cruise, M. Allen, J. Linderholm
Hutson, S.R., Magnoni, A., Beach, T., Terry, R.E., Dahlin, B.H., Schabel, M., 2009. and P. Reynolds” by Matthew Canti, Stephen Carter, Donald Davidson and Susan
Phosphate fractionation and spatial patterning in ancient ruins: a case study from Limbrey. Journal of Archaeological Science 33, 299–305.
Yucatan. Catena 78, 260–269. Maher, B.A., Alekseev, A., Alekseeva, T., 2002a. Variation of soil magnetism across the
Jans, M.M.E., Kars, H., Nielsen-Marsh, C.M., Smith, C.I., Nord, A.G., Arthur, P., 2002. In- Russian steppe: its significance for use of soil magnetism as a palaeorainfall proxy.
situ preservation of archaeological bone: a histological study within a multidisci- Quaternary Science Reviews 21, 1571–1576.
plinary approach. Archaeometry 44 (3), 343–352. Maher, B.A., Hu, M., Roberts, H.M., Wintle, A.G., 2002b. Holocene loess accumulation
Jenny, H., 1941. Factors of soil formation. McGraw-Hill, New York. and soil development at the western edge of the Chinese Loess Plateau:
Johnson, D.L., 2002. Darwin would be proud: Bioturbation, dynamic denudation, and the implications for magnetic proxies of paleorainfall. Quaternary Science Reviews
power of theory in science. Geoarchaeology An International Journal 17 (1), 7–40. 22, 445–451.
Johnson, D.L., Johnson, D.N., Moore, D.M., 2003. Deep and actively forming illuvial clay Maher, B.A., Alekseev, A.O., Alekseeva, T., 2003. Magnetic mineralogy of soils across the
in the reglolith and on bedrock. In: Dominguez, E., Mas, G.R., Cravero, F. (Eds.), 2001 Russian steppe: climatic dependence of pedogenic magnetite formation. Paleoge-
A Clay odyssey 12th International Clay conference July 22–28, 2001 Bahia Blance, ography, Palaeoclimatology, Palaeoecology 201, 321–341.
Argentina, pp. 205–210. Mandel, R.D., Bettis, E.A., 2001. Use and analysis of soils by archaeologists and
Johnson, D.L., Domier, J.E.J., Johnson, D.N., 2005a. Animating the biodynamics of soil geoscientists. In: Goldberg, P., Holliday, V.T., Ferring, C.R. (Eds.), Earth Sciences and
thickness using process vector analysis: a dynamic denudation approach to soil Archaeology. Kluwer Academic/Plenum, New York, pp. 173–204.
formation. Geomorphology 67 (1–2), 23–46. Marcelino, V., Cnudde, V., Vansteelandt, S., Carò, F., 2006. An evaluation of 2D-image
Johnson, D.L., Domier, J.E.J., Johnson, D.N., 2005b. Reflections on the nature of soil and analysis techniques for measuring soil microporosity. European Journal of Soil
its biomantle. Annals of the Association of American Geographers 95 (1), 11–31. Science 58 (1), 133–140.
Johnson, K.D., Terry, R.E., Jackson, M.W., Golden, C., 2007. Ancient soil resources of the Matthews, J.A., 1980. Some problems and implications of 14C dates from a podzol
Usumacinta River Region, Guatemala. Journal of Archaeological Science 34, buried beneath an end moraine at Haugabreen, southern Norway. Geografiska
117–1129. Annaler, Series A 62, 185–208.
Johnson, D.L., Johnson, D.N., Benn, D.W., Bettis, E.A., 2008. Deciphering complex soil/site Matthews, J.A., 1985. Radiocarbon dating of surface and buried soils: principles,
formation in sands. Geomorphology 101 (3), 484–496. problems and prospects. In: Richards, K.S., Arnet, R.R. (Eds.), Geomorphology and
Josephs, R.L., Bettis, E.A., 2003. Short contribution: a practical alternative to Kubiena soils. Allen and Unwin, London, pp. 269–288.
boxes for the collection of samples for micromorphological analysis. Geoarchaeo- Matthews, J.A., Dresser, P.Q., 1983. Intensive 14C dating of a buried paleosol. Geologiska
ogy 18 (5), 567–570. Föreningus I Stockholm Förhandlingar 105 (1), 59–63.
Kemp, R.A., 1985. Soil micromorphology and the quaternary. Quaternary Research Matthews, W., French, C.A.I., Lawrence, T., Cutler, D.F., Jones, M.K., 1997. Microstrati-
Association, Cambridge (Technical Guide No. 2). graphic traces of site formation processes and human activities. World Archaeology
Kemp, R.A., 2001. Pedogenic modification of loess: significance for palaeoclimatic 29, 281–308.
reconstructions. Earth Science Reviews 54 (1–3), 144–156. Mayer, J.H., 2002. Evaluating natural site formation processes in eolian dune sands: a
Knudson, K.J., Frink, L., Hoffman, B.W., Price, T.D., 2004. Chemical characterisation of case study from the Krmpotich Folsom Site, Killpecker Dunes, Wyoming. Journal of
Arctic soils: activity area analysis in contemporary Yup'ik fish camps using ICP-AES. Archaeological Science 29, 1199–1211.
Journal of Archaeological Science 31, 443–456. McCann, J.M., Woods, W.I., Meyer, D.W., 2001. Organic matter and anthrosols in
Lagerås, P., Sandgren, P., 1994. The use of mineral magnetic analyses in identifying Amazonia: interpreting the Amerindian legacy. In: Rees, R.M., Ball, B.C., Campbell,
Middle and Late Holocene agriculture – a study of peat profiles in Småland, C.D., Watson, C.A. (Eds.), Sustainable Management of Soil C.A.B. International,
Southern Sweden. Journal of Archaeological Science 21 (5), 687–697. Wallingford, pp. 180–189.
Langohr, R., 1993. Types of tree windthrow, their impact on the environment and their Middleton, W.D., 2004. Identifiying chemical activity residues on prehistoric house
importance for the understanding of archaeological excavation data. Helenium V. floors: a methodology and rationale for multi-elemental characterisation of a mild
XXXIII, 36–49. acid extract of anthropogenic sediments. Archaeometry 46 (1), 47–65.
134 H. Walkington / Earth-Science Reviews 103 (2010) 122–134

Morinaga, H., Inokuchi, H., Yamashita, H., Ono, A., Inada, T., 1999. Magnetic detection of Smith, N.J.H., 1980. Anthrosols and human carrying capacity in Amazonia. Annals of the
heated soils at Palaeolithic sites in Japan. Geoarchaeology: An International Journal Assoication of American Geographers 70, 553–566.
14 (5), 377–399. Smith, J., 1999. An Introduction to the magnetic properties of natural materials. In:
Nolin, L., Kramer, J.K.G., Newman, M.E., 1994. Detection of animal residues in humus Walden, J., Oldfield, F., Smith, J.P. (Eds.), Environmental magnetism: a practical
samples from a prehistoric site in the Lower Mackenzie River Valley, North West guide Technical Guide, No. 6 Quaternary Research Association, London, pp. 5–25.
Territories. Journal of Archaeological Science 21 (3), 403–412. Soil Survey Staff, 2010. Keys to Soil Taxonomy, 11th Edition. USDA-Natural Resources
Nortcliff, S.,1985. Soil Micromorphology. Progress in Physical Geography 9 (2), 281- Conservation Service, Washington DC.
290. Stafford, C.R., 2004. Modeling soil-geomorphic associations and Archaic stratigraphic
Peacock, E., Fant, D.W., 2002. Biomantle formation and artefact translocation in upland sequences in the lower Ohio River valley. Journal of Archaeological Science 31, 1053–1067.
sandy soils; an example from the Holly Springs National Forest, North Central Stoops, G., 2003. Guidelines for the analysis and description of soil and regolith thin
Mississippi, USA. Geoarchaeology An International Journal 17 (1), 91–114. sections. Soil Science Society of America, Madison.
Peters, C., Thompson, R., 1999. Supermagnetic enhancement, superparamagnetism and Sullivan, K.A., Kealhofer, L., 2004. Identifying activity areas in archaeological soils from a
archaeological soils. Geoarchaeology: An International Journal 14 (5), 401–413. colonial Virginia house lot using phytolith analysis and soil chemistry. Journal of
Peterson, S.C., Mohler, P.J., 2002. The archaeology of sandy soil sites: a new approach to Archaeological Science 31, 1659–1673.
the field evaluation of site integrity. North Carolina Archaeology 51, 98–122. Tanadarich, J.P., Darmody, R.G., Follmer, L.R., Johnson, D.L., 2002. History of soil science –
Rapp, G., Hill, C.L., 1998. Geoarchaeology, The Earth Science Approach to Archaeological historical development of soil and weathering profile concepts from Europe to the
Interpretation. Yale University Press, New Haven. 274pp. United States of America. Soil Science Society of America Journal 66 (2), 335–346.
Ripley, J.L., 1998. A geoarchaeological approach to using surface sites for paleaoenvir- Tang, Y., Jia, J., Xie, X., 2003. Records of magnetic properties in quaternary loess and its
onmental reconstructions. Geoarchaeology 13 (8), 793–818. palaeoclimatic significance. Quaternary International 108, 33–50.
Roberts, H.M., Wintle, A., Maher, B.A., Hu, M., 2001. Holocene loess accumulation rates Terry, R.E., Nelson, S.D., Carr, J., Parnell, J., Hardin, P.J., Jackson, M.W., Houston, S.D.,
in the western Loess Plateau, China, and a 2,500-year record of agricultural activity, 2000. Quantitative phosphorus measurement: a field test procedure for archaeo-
revealed by OSL dating. Holocene 11, 477–483. logical site analysis at Piedras Negras, Guatemala. Geoarchaeology: An Interna-
Robertson, K.M., Johnson, D.L., 2004. Vertical redistribution of pebbles by crayfish in tional Journal 15 (2), 151–166.
mollisol catenas of central Illinois. Soil Science 169 (11), 776–786. Theocharopoulos, S.P., Dalrymple, J.B., 1987. Experimental construction of illuviation
Romans, J.C.C., Robertson, L., 1983. The general effects of early agriculture on soils. In: cutans (Channel Argillans) with differing morphological and optical properties. Soil
Maxwell, G.S. (Ed.), The impact of Aerial Reconnaissance on Archaeology, Research Micromorphology 245–250.
Report No 49. Council for British Archaeology, London, pp. 136–141. Tipping, R., Carter, S., Johnston, D., 1994. Soil pollen and soil micromophological
Sampiettro, M.M., Vattuone, M.A., 2005. reconstruction of activity areas at a formative analyses of old ground surfaces on Biggar Common, Borders region, Scotland.
household in Northwest Argentina. Geoarchaeology: An International Journal 20 Journal of Archaeological Science 21 (3), 387–401.
(4), 337–354. Usai, M.R., 2001. Textural pedofeatures and pre Hadrians wall ploughed palaesols at
Sander, H., 2002. The porosity of tropical soils and implications for geomorphological Stanwix, Carlisle, Cumbria, UK. Journal of Archaeological Science 28, 541–543.
and pedogenetic processes and the movement of solutions within the weathering Valentine, K.W.G., Dalrymple, J.B., 1976a. Quaternary buried paleosols: a critical review.
cover. Catena 49 (1–2), 129–137. Quaternary Research 6, 209–222.
Sandor, J.A., 1992. Long term effects of prehistoric agriculture on soils: examples from New Valentine, K.W.G., Dalrymple, J.B., 1976b. The identification of a buried paleosol developed
Mexico and Peru. In: Holliday, V.T. (Ed.), Soils in Archaeology: Landscape Evolution in place at Pitstone, Buckinghamshire. Journal of Soil Science 27, 541–553.
and Human Occupation. Smithsonian Institution Press, Washington, pp. 217–245. Van Andel, T.H., 1998. Paleosols, Red Sediments and the Old Stone Age in Greece.
Sarris, A., Galaty, M.L., Yerkes, R.W., Parkinson, W.A., Gyucha, A., Billingsley, D.M., Tate, Geoarchaeology: An International Journal 13 (4), 361–390.
R., 2004. Geophysical prospection and soil chemistry in the Early Copper Age Van Nest, J., 2002. The Good earthworm: how natural processes preserve upland
settlement of Véstő-Bikeri, Southeastern Hungary. Journal of Archaeological archaic archaeological sites of Western Illinois, USA. Geoarchaeology An Interna-
Science 31, 927–939. tional Journal 17 (1), 53–90.
Schaetzl, R., Anderson, S., 2005. Soils: Genesis and Geomorphology. Cambridge Van Vliet-Lanoë, B., 1985. Frost effects in soils. In: Boardman, J. (Ed.), Soils and
University Press, Cambridge. 817 pp. Quaternary Landscape Evolution. John Wiley, Chichester, UK, pp. 117–158.
Schnitzer, M., Calderoni, G., 1985. Some chemical characteristics of paleosol humic Van Vliet-Lanoë, B., 1998. Frost and soils: implications for paleosols, paleoclimates and
acids. Chemical Geology 53, 175–184. stratigraphy. Catena 34 (1–2), 157–183.
Schuldenrein, J., Wright, R.P., Mughal, M.R., Khan, M.A., 2004. Landscapes, soils, and mound Van Vliet-Lanoë, B., Coutard, J.P., 1984. Structures caused by repeated freezing and
histories of the Upper Indus Valley, Pakistan: new insights on the Holocence thawing in various loamy sediments: a comparison of active, fossil and
environments near ancient Harappa. Journal of Archaeological Science 31, 777–797. experimental data. Earth Surface Processes and Landforms 9, 553–565.
Scott-Jackson, J.E., Walkington, H., 2005. Methodological issues raised by laser particle Vanniere, B., Bossuet, G., Walter-Simonnet, A.-V., Gauthier, E., Barral, P., Petit, C.,
size analysis of deposits mapped as clay-with-flints from the Palaeolithic site of Buatier, M., Daubigney, A., 2003. Land use change, soil erosion and alluvial dynamic
Dickett's Field, Yarnhams Farm, Hampshire, UK. Journal of Archaeological Science in the lower Doubs Valley over the 1st millennium AD (Neublans, Jura, France).
32 (7), 969–980. Journal of Archaeological Science 30, 1283–1299.
Shahack-Gross, R., Marshall, F., Weiner, S., 2003. Geo-ethnoarchaeology of pastoral Viklund, K., 1998. Cereals, weeds and crop processing in Iron age Sweden.
sites: the identification of livestock enclosures in abandoned Maasai settlements. Methodological and Interpretive Aspects of Archaeological Evidence. Archaeology
Journal of Archaeological Science 30, 439–459. and Environment 14, Umea.
Shahack-Gross, Albert, R.M., Gilboa, A., Nagar-Hilman, O., Sharon, I., Weiner, S., 2005. Wagner, D.P., McAvoy, J.M., 2004. Pedoarchaeology of Cactus Hill, a Sandy Paleoindian
Geoarchaeology in an urban context: the use of space in a Phoenician monumental Site in Southeastern Virginia, U.S.A. Geoarchaeology: An International Journal 19
building at Tel Dor (Israel). Journal of Archaeological Science 32, 1417–1431. (4), 297–322.
Simonson, R.W., 1959. Outlining of a generalized theory of soil genesis. Soil Science Wandsneider, L., 1988. Experimental investigation of the effect of dune processes on
Society of America Proceedings 23, 152–156. archaeological remains. American Archaeology 7, 18–29.
Simonson, R.W., 1978. A multiple-process model of soil genesis. In: Mahaney, W.C. Waters, M.R., 1992. Principles of Geoarchaeology a North American Perspective.
(Ed.), Quaternary Soils, 3rd symposium on Quaternary research. May 21–23, 1976, University of Arizona Press, Tucson. 398 pp.
Toronto, Canada, pp. 1–25. Webb, E.A., Schwarcz, H.P., Healy, P.F., 2004. Detection of ancient maize in lowland
Simpson, I.A., 1997. Relict properties of anthropogenic deep topsoils as indicators of Maya soils using stable carbon isotopes: evidence from Caracol, Belize. Journal of
infield management in Marwick, West Mainland, Orkney. Journal of Archaeological Archaeological Science 31, 1039–1052.
Science 24 (4), 365–380. West, A.G., Bond, W.J., Midgley, J.J., 2000. Soil carbon isotopes reveal ancient grassland
Simpson, I.A., Bryant, R.G., Trevaabak, U., 1998a. Relict soils and early arable land under forest in Hluhlwe, KwaZulu, Natal. South African Journal of Science 16, 252–254.
management in Lofoten, Norway. Journal of Archaeological Science 25 (12), Wilkinson, T., 1999. Holocene Valley fills of Southern Turkey and northwestern Syria:
1185–1198. recent geoarchaeological contibutions. Quaternary Science Reviews 18, 555–571.
Simpson, I.A., Dockrill, S.J., Bull, I.D., Evershed, R.P., 1998b. Early anthropogenic soil Wilkinson, T., 2005. Soil erosion and valley fills in the Yemen highlands and southern
formation at Tofts Ness, Sanday, Orkney. Journal of Archaeological Science 25 (8), Turkey: integrating settlement, geoarchaeology and climate change. Geoarchaeol-
729–746. ogy: An International Journal 20 (2), 169–192.
Simpson, I.A., Guðmundsson, G., Thomson, A.M., Cluett, J., 2004. Assessing the role of Wilson, C., Simpson, I.A., Currie, E.J., 2002. Soil management in pre-Hispanic raised field
winter grazing in historic land degradation, Mývatnssveit, Northeast Iceland. systems: micromorphological evidence from Hacienda Zuleta, Ecuador. Geoarch-
Geoarchaeology: An International Journal 19 (5), 471–502. aeology 17 (3), 261–283.
Simpson, I.A., Barrett, J.H., Milek, K.B., 2005. Interpreting the Viking Age to Medieval Woods, W.I., McCann, J.M., 1999. The Anthropogenic origin and persistence of
Period transition in Norse Orkney through Cultural Soil and Sediment Analyses. Amazonian Dark Earths. Conference on Latin Americanist Geographers 25, 7–14.
Geoarchaeology: An International Journal 20 (4), 355–377. Yaalon, D.H., 1986. Paleosols in the mainstream pedological literature: a review of some
Simpson, I.A., Guttmann, E.B., Cluett, J., Shepherd, A., 2006. Characterising anthropic recent texts. Progress in Physical Geography 10 (4), 587–593.
sediments in North European Neolithic settlements: an assessment from Skara
Brae, Orkney. Geoarchaeology: An International Journal 21 (3), 221–235.