Since the first prehistoric people started to dig for stone to make implements, rather than pick up loose material, humans have modified the landscape through excavation of rock and soil, generation of waste and creation of artificial ground. In Great Britain over the past 200 years, people have excavated, moved and built up the equivalent of at least six times the volume of Ben Nevis. It is estimated that the worldwide deliberate annual shift of sediment by human activity is 57 000 Mt (million tonnes) and exceeds that of transport by rivers to the oceans (22 000 Mt) almost by a factor of three. Humans sculpt and transform the landscape through the physical modification of the shape and properties of the ground. As such, humans are geological and geomorphological agents and the dominant factor in landscape evolution through settlement and widespread industrialization and urbanization. The most significant impact of this has been since the onset of the Industrial Revolution in the eighteenth century, coincident with increased release of greenhouse gases to the atmosphere. The anthropogenic sedimentological record, therefore, provides a marker on which to characterize the Anthropocene.
Since the first prehistoric people started to dig for stone to make implements, rather than pick up loose material, humans have modified the landscape through excavation of rock and soil, generation of waste and creation of artificial ground. It is estimated that the worldwide deliberate annual shift of material by human activity is 57 000 Mt (million tonnes) and exceeds that of transport by rivers to the oceans (22 000 Mt) almost by a factor of three . Humans sculpt and transform the landscape through the physical modification of the shape and properties of the ground surface and subsurface. As such, humans are geological and geomorphological agents and the dominant factor in landscape evolution in the Anthropocene [2,3]. The magnitude of impact (quantity and spatial extent) of material moved and its rate correlates with increasing population. The exploitation of the landscape and its subsurface to meet the needs of society is driven by changes in socioeconomic, technological, political and cultural parameters.
The geological and geomorphological impact of these changes is reflected in the deposition of made ground, removal of material through excavation, enhanced denudation through agricultural activity and the disruption of natural sediment transfer through dam construction, for example. The former two processes represent the deliberate actions of humans to extract, transfer, re-use or discard rock, soil and man-made materials. This contrasts with the latter two processes, where the impacts on sediments and sediment transfer are the unintentional consequence of anthropogenic processes where movement of sediment was not the primary objective. Anthropogenic processes where rock or soil is deliberately excavated, transported and deposited by people include: mining (subsurface and opencast), construction for settlement, industry, burial and defensive purposes, processing of metal ore, waste generation, and canal, road and railway infrastructure construction. Unintentional activity relates to a combination of aeolian, hydrological and slope erosional processes related to soil loss and transport through agricultural modification of the landscape for pasture and arable land. Although both intentional and unintentional processes result in significant sedimentary markers on which the Anthropocene may be characterized, it is the deliberate, anthropogenic processes on which this paper focuses.
Widespread industrial activity and urbanization accelerated during the Industrial Revolution and are associated with increased input of carbon dioxide (CO2) and methane (CH4) into the atmosphere. It is proposed that this may represent a significant atmospheric marker defining the Anthropocene epoch [2,4]. However, the deliberate modification of the landscape to support human habitation and culture through resource exploitation, construction, demolition and deposition of wastes provides a sedimentological basis for defining the Anthropocene. Such sedimentological anthropogenic processes and their impacts are most significant during and after the Industrial Revolution of the late eighteenth century (figure 1). This period represents significant population growth, industrialization and large-scale exploitation of subsurface resources. This subsurface signature in the sedimentary record may provide a physical basis on which to characterize the Anthropocene. However, on a smaller scale, landscape-transforming anthropogenic processes occurred before the eighteenth century and the onset of widespread industrialization.
Evidence for burning in upland areas and the discovery of Mesolithic middens, especially near the coasts of Great Britain, suggests that landscape modification may have begun at this time. However, its spatial extent and the volume of material moved are likely to have been limited in Great Britain. Evidence of significant deliberate landscape modification is recognized as early as the Neolithic as people transformed their cultural way of life from hunter–gatherers to farming and settlement. Trade of goods may have also taken place at this time, reflecting early exploitation of subsurface resources of quality flint for tool making. Deliberate, landscape transformation through mineral exploitation, processing and waste generation gathered pace during the Bronze Age. It was around this time that transformation of the landscape increased and humans became geological and geomorphological agents through the deliberate removal, transport and placement of sediment. This deliberate, anthropogenic modification of the sedimentological record may provide a marker for the onset of the Anthropocene, although on this basis, its onset is likely to be diachronous.
The Roman occupation of Great Britain marked a time of population centralization in towns and cities. Subsequent successive phases of urban expansion acted as the driver for the increased landscape modification within these towns and cities and on the surrounding landscape that provided the resources to support their development. Although the timing of this is likely to be diachronous, it may be possible to define the Anthropocene on the basis of anthropogenic geological and geomorphological landscape transformation around this time.
Human activity and land-use change have left an imprint above and below the land in the form of artificial ground. Artificial ground includes areas where material is known to have been placed by humans on the pre-existing natural land surface and areas where the pre-existing land surface is known to have been excavated. It can be a ground hazard, an archaeological heritage deposit and a potential mineral resource where economic conditions permit. Deposited in the shallow subsurface through historical and current anthropogenic activity, it forms part of a shallow zone of human interaction and has become part of the geological record from which the Anthropocene can be characterized. Owing to its deposition in predominantly terrestrial environments, the long-term geological preservation potential of artificial ground is uncertain.
By understanding the physical impacts of historical and current changes in urban and rural land use, it may be possible to predict the resilience and future response of the Earth’s shallow subsurface to environmental change.
2. Humans as geological and geomorphological agents and the shallow zone of human interaction
Landscape is a complex concept that can include topography, landform, an object, a way of seeing the world and an inhabited space . The modern landscape represents the surface expression of the underlying geology and the geological and anthropogenic processes that have transformed it. Landscape-forming processes in Great Britain reflect a combination of natural and anthropogenic systems that remove, transport and emplace sediment. This flow of materials and their inclusion in the subsurface geological record is one of the defining processes that characterize the landscape that we see around us today.
Rapid population growth and the exploitation of the Earth’s resources through urbanization, industrial and agricultural activity have led to humans becoming one of the most significant factors in the evolution of the Earth’s landscape. As such, humans are a geological and geomorphological agent [1,9–12]. The impacts of anthropogenic transformation of the landscape through the transfer of quantities of rock and soil and the legacy it has left above and below the ground as artificial ground may be used as an indicator of human activity with which to characterize the Anthropocene [2,3].
While the ability of Homo sapiens to deliberately transform their environment is not unique in the animal kingdom, the magnitude of their impact (principally quantities of rock and soil transferred) is significant. Douglas & Lawson  estimate that 972 Mt of Earth material are deliberately moved by humans in Great Britain each year. This is comparable with the capacity of annelid earthworms, for example, to process soil after ingestion. Darwin  estimated that rates of soil processing in earthworms are as much as 18.12 tonnes per acre per year (equivalent to an estimate of 991 Mt per year in Great Britain).
In Great Britain, the geomorphological and geological resources of the Earth have provided the raw materials to support the expansion of human population. These needs include the provision of raw materials for shelter, energy, technology and the space for the disposal of wastes. Population growth, industrial expansion and urbanization are, therefore, intimately linked to the direct impact of human activity on the landscape. Through history, there have been pulses of intense landscape transformation taking place over short periods of time, punctuated by wars, economic depressions, famine and disease. Evidence of this legacy has the potential to be preserved as artificial ground either above or below the land surface.
Pre-industrial (broadly pre-eighteenth-century) human activity associated with the transition of human society towards settlement, agriculture and localized industrial activity evolved in Great Britain from the Neolithic. Localized industrial and settlement activity included subsurface Neolithic flint workings at Grimes Graves in Norfolk , quarries for stone axe factories in Langdale (NW England) and Bronze Age copper mining on the Great Orme  and Parys Mountain, Anglesey, North Wales . The onset of such activities resulted in the extraction, transport and placement of rock and soil as artificial ground. Where people settled they also deposited wastes, so that middens form part of the anthropogenic strata comprising artificial ground. These types of activities represented the first examples of humans contributing to the flow of geological material, and their role as geological and geomorphological agents in Great Britain.
The Iron Age saw the expansion of agricultural communities from their origins in the Neolithic, and an emerging culture working and smelting iron ore for weapons and implements and constructing defensive hill forts. There are numerous mining and smelting sites, such as those in the Forest of Dean and Northamptonshire in England, that show evidence of landscape excavation and waste deposition . This activity created a need for charcoal to smelt the ore, leading to deforestation and the start of human influence on soil erosion.
Roman Britain saw an increase in mineral production for various commodities, including copper, salt, coal and lead [15,18]. At the same time, major building projects using large amounts of quarried rock and burnt lime for concrete were undertaken as Roman populations became centralized into urban centres, forming the foundations of many future towns and cities . Such activity formed the first significant peak in the construction of the British built environment as towns and cities became the focus for manufacturing, processing and trade. Subsequent phases of urban growth, construction and demolition created an often thick ‘urban deposit’ including construction materials, foundations and wastes in middens. Urban growth rates fluctuated during the Anglo-Saxon, Anglo-Scandinavian and Medieval times, reaching a peak of new town construction in the thirteenth century , but mining continued for various metals and coal. During pre-industrial times, the relative scale of surface and subsurface impacts on the landscape may have been significant, but the geographic extent was generally limited (table 1).
Wider geographic impacts were associated with intensive industrial activity, settlement and urbanization, beginning in the late eighteenth century. Steffen et al.  refer to this era as the Industrial Era, stage 1 of the Anthropocene, extending from ca1800 to 1945. They recognize this time as a major phase of human influence on the Earth system, significantly impacting the atmosphere through increased inputs of CO2 and CH4. However, physical transformation of the landscape through anthropogenic modification of the surface and subsurface of the Earth was also significant. Mining activities, urbanization and waste production resulted in large-magnitude impacts through increased volumes of Earth material extracted, processed and deposited over increasingly short periods of time compared with the pre-eighteenth century. This resulted in the large-scale creation of artificial ground. The intensity of human activity, resource exploitation, material consumption and landscape transformation varies through time, although the relative scale of impacts of that activity may be greater over shorter periods as populations expand (figure 1). For example, coal production over the 350 years between 1500 and 1850 was around 2895 Mt . Later, an equivalent amount of coal was produced in only 23 years between 1860 and 1883. A coal mining peak in Great Britain between 1910 and 1920 witnessed the production of the same volume in just 10 years  according to BGS mineral statistics .
Deposition of geological materials derived from anthropogenic activity includes the emplacement of construction materials and the disposal of wastes. While deposition may occur close to the source of extraction, construction of the built environment often requires the transport and net import of materials from a wide catchment. Major urban areas might, therefore, be thought of as sinks for the deposition of geological materials that are either directly won from the ground or processed for re-use in construction (e.g. manufacture of concrete and bricks). Anthropogenic activity including transport and use of construction material can, therefore, create an urban ‘deposit’ through inflow of material into the built environment . Although centralized urban construction for residential, commercial and industrial purposes focuses deposition of transferred material as artificial ground, its impact on rural environments is also significant. Infrastructure and people in urban areas rely on resources transported in to them to sustain their growth. Transport of agricultural products and construction materials among others is required, along with the road, rail and canal transport routes to carry them either from other parts of Great Britain or increasingly from international sources via ports and airports. Similarly, just as urban areas receive a net inflow of resources to support them, they are also net generators of waste materials. These materials may be buried in landfill sites or recycled and re-used in construction where the ‘sediment’ is fit for that purpose. In all cases, the deposition of processed natural rock or soil raw materials results in the creation of artificial ground and constitutes an anthropogenic feature of the Anthropocene.
The political, socioeconomic and technological drivers that accelerated Britain’s industrial growth transformed the pattern of human habitation in the eighteenth and nineteenth centuries. Previously, localized habitation and industry were generally restricted by the availability of energy sources such as water and wind. Industrialization resulted in a rapid transformation to city living and centralized human occupation that began as far back as the Roman period. Globally, in 1890 approximately 200 million people lived in cities compared with approximately 3 billion in 2001 . In the UK alone, nearly 80 per cent of people lived in an urban area in 2001 .
It is estimated that by the Domesday survey of 1086, lowland Britain was already densely populated. Deforestation as a result of this meant that woodland comprised just 5 per cent of the landscape of eastern England and 15 per cent of England as a whole . The agricultural impacts of soil losses through processes such as tilling following tree clearances are significant. Some agricultural practices including tilling, mechanization and increased field size increase the vulnerability of soils to aeolian and hydrological erosion. The erodibility of soils depends largely on the lithological composition of the soil. It is estimated that, in the developed world, annual soil losses (denudation rate) from land classified as cropland are of the order of 1.0–3.5 kg m−2 for land classified as pasture . In some parts of northeastern and eastern England, landmass areas classified as arable are susceptible to soil erosion rates in the region of 2.1–4.4 kg m−2 through aeolian erosion alone .
Global anthropogenic denudation rates have exceeded those of natural, deep-time denudation since the first millennium . Global, deep-time denudation rates have fluctuated through the Phanaerozoic, ranging from approximately 10 m per million years (m Myr−1) in the Middle Triassic to over 60 m Myr−1 during the Pliocene, with a mean average of 25 m Myr−1 . When converted to geological time scales, denudation from anthropogenic cropland tillage alone is estimated to be of the order of 680 and 1400 m Myr−1 in developed and undeveloped countries, respectively [23,24]. If the average denudation rate from agricultural activities is approximately 643 m Myr−1 , then the rate of agricultural anthropogenic denudation is approximately 28 times that of the deep-time mean denudation rate.
Globally, Hooke  estimates that, in the year 2000, approximately 21 tonnes of rock and soil were moved per capita through combined agricultural and construction activities. The physical transformation of the Earth through human-induced processes such as transfer of rock and soil or indirect erosion through tilling defines a shallow zone of human interaction in the geosphere. The end products of anthropogenic processes (either a sedimentary deposit or an erosional feature) provide a physical marker within the Anthropocene.
3. Anthropogenic processes
Evidence of anthropogenic processes is recorded in the landscape and its subsurface as either constructional or erosional features. Rock, soil, anthropogenic (processed or manufactured) deposits or a mixture can be deposited and used in either a controlled engineered or an uncontrolled manner. Commonly, deposition results in a constructional landform, an engineered structure or, where the landform is poorly defined on the surface, a diffuse sedimentary deposit in the ground. Such deposits are referred to as made ground (see §4b). Equally, anthropogenic processes may be expressed as excavations, representing natural material that has been removed through extraction. Such excavations are referred to as worked ground. The material flux associated with anthropogenic processes of removal, transport and deposition are similar to those geological processes of erosion, transport and deposition.
(a) Habitation and infrastructure
The remains from habitation buried in the ground are recorded by archaeologists with a well-defined historical chronology. However, for the past 40 years or so, only the major features such as earthworks have been recorded by geologists as a part of artificial ground research. Many long-established cities have considerable thicknesses of artificial ground, caused by successive phases of habitation building on the remains of earlier development. This can result in made ground many metres high composed of building rubble and waste or, in large cities, widespread made ground that has a poorly defined margin and is very heterogeneous. This complexity of artificial ground beneath cities began with the onset of urban centralization during the Roman period  when buildings, transport infrastructure, irrigation and municipal waste tips were constructed. Successive phases of development have added to, or in some cases re-used and recycled, this artificial ground, leaving a complex ‘stratigraphy’ of deposits, including drains, middens, pits, cellars, foundations and trenches among other features. For example, in industrial Manchester and Salford, NW England, this has resulted in up to 10 m of artificial ground. In the heritage city of York, NE England, equivalent deposits, reflecting over 2000 years of occupation, are up to 8 m thick .
From an early age, the infrastructure that supports settlement and habitation included the construction of irrigation systems followed by canals, roads, railways and tunnels for transport. These have been excavated beneath the land surface, cut into the land or banked upon the natural ground surface, forming worked and made ground. From the Industrial Revolution onwards, the scale of these activities has increased and the volumes of material now moved during construction are considerable. For example, the Channel Tunnel, linking Great Britain and mainland Europe, produced 20 Mt of rock spoil, which was placed at the foot of Shakespeare Cliff, forming 40 ha (hectares) of new land at Samphire Hoe in Kent, SE England . The CrossRail railway scheme linking east and west London produced 5 Mt of material from the rail link excavations, which was shipped down-river to form 600 ha of made ground at Wallasea Island, Essex, eastern England . Construction of an extension of the M6 motorway in the English Midlands resulted in the movement of approximately 15 Mt of material through excavation and embankment construction .
Currently, the amount of material moved for road and infrastructure construction in Great Britain is probably about 155 Mt a year based on year 2000 figures . Historical made ground is most likely to be unengineered tipped fill. However, modern deposits for road and infrastructure construction are compacted and engineered for strength and performance; the majority of this built-up material is of local or moderately local origin, though in coastal areas it may have been imported, sometimes as ship ballast from abroad. The volume of material moved for the construction of canals, railways, roads, motorways and tunnels and for building construction is significant. In 1922, Sherlock estimated the volume of material moved for the construction of canals, railways, tunnels, roads and building construction as 4507 million cubic yards or about 3.5 km3 . Since then, significant urban expansion and motorway construction have taken place, moving approximately 1500 Mt in the last 10 years and possibly 11 000 Mt since 1922, a volume of around 8–9 km3.
(b) Natural resource exploitation
The human need for minerals, energy and building materials has resulted in widespread mining and quarrying. The relative importance of cultural, socio-economic, political and environmental factors as drivers for the growth of populations and their use of natural resources changes through time. Mineral veins at the surface have been worked to leave opencast excavations in the landscape and spoil heaps as made ground. Quarrying, first for building stone, then later for brick clay, aggregate and cement manufacture, has left large-scale areas of worked ground. Quarries for igneous and metamorphic rocks are visible in areas such as North Wales, the Lake District and the Midland Valley of Scotland. Technology such as hushing and ground sluicing, to locate mineral veins, is first seen in the Roman period, harnessing natural forces to aid discovery and production. To deliver the large quantities of water required to the mine head, large industrial aqueducts and hushing tanks were constructed. This system was used at the Roman gold mine Dolaucothi in southern Wales, where widespread archaeological evidence for these processes has been recorded . Opencast mineral working, especially for coal and ironstone, has moved considerable volumes of overburden. This may reach 18 or 20 times the volume of the mineral extracted. In some areas of Great Britain, where the Coal Sequences come to surface, up to 20 per cent of the land has been turned over in this manner. Some of these workings reach down to depths of 80 m or more, and the deepest, the former Westfield site in Scotland, to 215 m .
Subsurface working has resulted in collapses and open stopes, creating subsidence that lowers the natural ground surface. This subsidence is effectively ‘hidden’ artificial ground. Underground coal workings were initially undertaken using pillar and stall extraction techniques, leaving earlier random and then later regular patterns of galleries with supporting pillars. Where underground workings have been infilled for ground stabilization, or where equipment and infrastructure have been left underground, artificial ground is created. Industrial impacts on the shallow subsurface are widespread and arise not only from the physical infrastructure associated with the mining industry but also from the chemical and biological processes that underpin these activities. The examination of 9 m deep alluvial deposits on the banks of the Ouse at North Street, York, revealed the presence of lead contamination arising from relatively remote Yorkshire Dales mining activity between the ninth and thirteenth centuries AD. Contamination was sourced from mining and smelting activities and contaminants transported by the River Ouse. There were also local effects from consumption, including use, wear and disposal .
Using BGS figures for extraction since 1850, the total UK mined (onshore and undersea) and opencast coal is around 24 700 Mt or a volume of around 19.5 km3 of extraction. The waste from this adds another 6800 Mt, with a solid volume of around 3.1 km3 and a bulked-up volume of around 5 km3. UK production of iron ore from 1850 to 1973 (figure 1) was about 1582 Mt. A bulk density of 2.59 g cm−3 suggests an extracted volume of about 0.6 km3 plus 1.8 km3 of overburden. A significant amount of the ore volume has been turned into furnace slag and waste tips. These are just a few indicators of the magnitude of impact and landscape modification by humans creating artificial ground and subsidence. Applying similar calculations to all of the major commodities listed in the BGS mineral statistics for the UK gives a total volume of around 38.5 km3 of material (mineral, overburden and waste) moved by humans for mineral extraction since 1850. This is equivalent to moving more than four times the volume of Ben Nevis in the past 160 years.
From early in the twentieth century, longwall mining of coal became the standard extraction technique, made possible by technology allowing roof support and ventilation . In the longwall workings, panels of coal were extracted between parallel tunnels dug out from a service or haulage tunnel. This technique left less of the resource underground, but caused largely predictable subsidence at the surface. The mechanized extraction also generated more waste than manual digging.
Some minerals such as salt are soluble and were originally gathered from natural brine springs. With industrialization, pumping of the wild brine allowed large amounts of production. It also caused severe subsidence in places such as Northwich and Droitwich in central England . The brine run subsidence features can extend for 10 km or so from the salt works pumping them. Many square kilometres of the Cheshire salt area, NW England, are affected by this sort of subsidence and similar features; brine extraction and subsidence also affect land near Preesall (Lancashire, NW England) and Teesside (NE England). Such features are now integral parts of the landscape. For example, lakes or ‘meres’ now occupy salt subsidence features in Cheshire.
Similarly, the areas and volumes of subsided land caused by coal mining are considerable. In Great Britain, it is not uncommon for between 1 and 3 m of subsidence to have occurred due to mining operations. Toll Bar near Doncaster, NE England, is an example of this; a comparison of benchmarks here shows up to 3.1 m of subsidence between 1948 and 2007. Elsewhere, up to 13 m of subsidence has occurred at Silverdale near Wigan, NW England (D. Wilshaw 2010, personal communication). From mining literature [36,37], it is likely that there is a range of values from 50 to 90 per cent of the mined volume manifesting itself as surface subsidence. BGS mineral figures for deep coal mining since 1900 (assumed to be mainly longwall mining) indicate 18 151 Mt of coal extraction. With a density of 1.26 g cm−3  and a waste extraction of 35 per cent, this equates with a surface subsidence volume of between 9.7 and 19.4 km3 depending on what percentage of subsidence manifests itself at the surface. It must be noted that a proportion of this subsidence will be under the sea.
(c) Natural resource processing and wastes
Deep coal mining generates about 35 per cent waste, which is deposited as spoil heaps, which form one of the most significant anthropogenic landforms in Great Britain. Information from Ordnance Survey maps shows some spoil heaps covering between 0.5 and 1 km2 (e.g. Gascoigne Wood Mine, NE England), containing millions of tonnes of waste. Most mineral processing generates waste and some, such as iron smelting and oil shale processing, generate very large volumes compared with the end product. Artificial ground comprising furnace slag covers large areas around many British towns and cities. The form of these waste heaps varies from small, randomly tipped heaps near historical mines and smelters through more organized tips built out from mine or works railways or conveyors. Modern tips can extend over many hundreds of hectares, can be tens of metres high and, unlike their earlier counterparts, can be heavily engineered for stability. In coastal areas, much of the waste material has been used for land reclamation (e.g. Teesside, NE England, and Port Talbot, South Wales), where reclaimed land several kilometres across has been built up by furnace wastes and other spoil . The tonnages and volumes of waste products produced in the UK each year are very large, but some of the materials can be re-used, often being re-deposited in the urban environment.
Domestic and industrial waste tips are a type of artificial ground and commonly occupy previously excavated holes in the ground or sit upon the ground if there are no convenient workings. For example, in NE England much of Ferrybridge Power Station’s fly ash is deposited in settling ponds constructed in the former sand and gravel pits of the Aire Valley, North Yorkshire. Old waste tips commonly fill quarries and small river valleys and were completely uncontrolled in their content . Modern waste disposal sites by comparison are licensed and have facilities for the interception and removal of gas and leachate. In 1995, municipal and household wastes were 25.89 Mt, and 88.6 per cent of this went to landfill (22.9 Mt), with only 6.4 per cent being recycled . By 2005/2006, there were 35.1 Mt of municipal waste, of which 64 per cent (22.5 Mt) went to landfill, 8 per cent was incinerated and 27 per cent was recycled . Demolition waste is also being recycled and now approximately half (about 42 Mt in 2005) is re-used . Prior to about 2000, most went into landfill sites. The volumes of waste deposited as artificial ground are difficult to determine owing to the differing ways in which it has been classified. However, Department for Environment, Food and Rural Affairs (DEFRA) figures show that, from 1988 to 2008, the total amount of landfill material was about 800 Mt, equivalent to about 0.8 km3 of material in 10 years.
4. Anthropogenic activity in the geological record: artificial ground classification and characterization
(a) Natural and anthropogenic landscape evolution
Human and natural processes rarely operate in temporal or spatial isolation. Anthropogenic processes contribute to the evolution of the landscape through the production of artificial ground, including archaeological remains. This includes areas where human activity has modified the landscape through removal or placement of rock, soil and waste material. The type of excavation and composition of anthropogenic material reflect the process that emplaced it and its origin. Its composition can be extremely variable both laterally and vertically, representing rapid land-use change in one locality. Stratigraphically, artificial ground can be interpreted as sedimentary deposits or excavations representing the human geological record during the Anthropocene.
(b) Artificial ground classification
It was not until the 1960s that artificial ground was shown on geological maps published by the BGS , and not until the 1990s that artificial ground mapping became a routine aspect of geological surveying [44,45]. Artificial ground is recognized as a potential ground hazard owing to the variability of its composition, ground stability (including landslides) and its association with potentially contaminative land uses [46,47]. However, it also represents the ‘geological’ evidence for historical anthropogenic activity and forms part of the shallow zone of human interaction in the subsurface. Where socioeconomic conditions are suitable, some types of artificial ground may be a resource (former metalliferous waste tips, for example). Artificial ground may also enhance ecosystems and the services they provide, for example, through the creation of ecological habitats in areas of historical mineral extraction.
Artificial ground characterization requires the identification of a diagnostic landform (morphology) and where possible its physical, sedimentological and lithological (compositional) characteristics . In general, a combination of these characteristics is required to interpret the genesis of artificial ground, though they may not always be available. For example, a landform may be well defined at the surface but its subsurface properties and thickness may be unknown. Conversely, subsurface data from boreholes and excavations may be available, but there may be no surface landform associated with it . Modern 1 : 10 000 and 1 : 50 000 scale British geological maps use a fivefold classification scheme in their depiction of artificial ground (table 2 and figure 2). This artificial ground classification is based on a morpho-stratigraphic approach with an emphasis on the landform and the anthropogenic process that created it. In this scheme, made ground includes engineered fill. Landforms are identified through a combination of observation and the appraisal of spatial data sources, including aerial photographs, topographical maps, digital elevation models and ground investigation data (figure 3). The combined use of recent spatial data and multiple generations of legacy data allows historical land-use change to be considered and the most appropriate class of artificial ground to be chosen. However, with the exception of infilled ground, the scheme does not account for different phases of anthropogenic activity that may be represented at any one location.
In response to a continuing increase in the recognition of artificial ground as a key component of the human zone of interaction [47,51], an enhanced classification scheme for artificial ground in two and three dimensions has been devised [49,50]. This extensible scheme is structured as a three-tier hierarchy, using class, type and unitto describe in progressively more detail the origin and landform of the deposit or excavation (figure 4). Where multiple sources or generations of artificial ground exist, the temporal and aerial extent from each source may be recorded to capture the development of the artificial ground as a factor in anthropogenic landscape evolution (figure 5). Thematic surveys, including a detailed study of central Liverpool, have applied the enhanced scheme to develop an integrated anthropogenic and natural landscape evolution model (figure 6 and table 3). The scheme is designed to interface with complementary schemes, including the National Land Use Database, which contains information on previously developed land in England . Parallel schemes for the classification of urban soils offer the potential for further integration and the development of a comprehensive reference framework for urban environmental research . The enhanced artificial ground classification scheme provides a basis for the quantification of anthropogenic activity within the Anthropocene.
(c) Artificial ground distribution
The gross distribution of mapped artificial ground in Great Britain is a function of the date when individual areas (1 : 50 000 scale geological map sheets) were last surveyed (figure 7). Although geological data exist for almost every sheet, only those surveyed since the 1960s (approx. two-thirds of Great Britain) show any significant artificial ground. Based on an assessment of available geological map data, approximately 1.4 per cent of mainland Britain is covered by artificial ground (table 4).
The extent of mapped artificial ground is greatest in urban conurbations, where the landscape has been affected by frequent phases of human activity. For example, in Manchester artificial ground covers almost 18 per cent of the land and includes colliery spoil, infilled brick and gravel pits, industrial waste and infilled river valleys . Rural areas, commonly the source of much of the material destined for accumulation in urban and industrial conurbations, show less than 1 per cent artificial ground coverage. For example, artificial ground in Saxmundham in rural Suffolk represents only 0.7 per cent of the sample area, comprising infrastructure, coastal defences and mineral extraction sites that have provided aggregates for surrounding urban areas. Composite areas such as the Midland Valley of Scotland and the wider London area combine both the source and accumulation sites for artificial deposits and show an average coverage of approximately 7.2 per cent by area.
The figures presented here for the extent of artificial ground are an underestimate of their probable total coverage. The limitations of conventional geological surveying and map production (scale restrictions and cartographic generalization) result in an under-representation of artificial ground. Similarly, rapid land-use change, especially in urban areas, means that recent artificial ground may be unrecorded.
Recent advances in three-dimensional geological modelling of the shallow subsurface have enabled the development of the three-dimensional characterization and classification of artificial ground. Three-dimensional geological modelling methodologies using GSI3DTM software  have been applied to artificial ground to quantify and visualize anthropogenic landscape evolution in Manchester and Liverpool, NW England . Coverage of three-dimensional geological models in Great Britain is increasing but is not yet as extensive as two-dimensional geological maps. Extending the two-dimensional classification into three-dimensional is an essential tool in characterizing the legacy of anthropogenic impacts, including concealed anthropogenic events in the subsurface. Three-dimensional geological models capture the vertical sequence and volume of individual deposits and excavations, providing a three-dimensional framework for attribution with age and composition information. This is especially important in areas with specific historic activities that may be obscured beneath younger made ground or natural deposits. Figure 8 shows the thickness of made ground resulting from industrial land-use change over approximately 160 years in Trafford Park, Salford, in NW England. This is derived from a three-dimensional geological model based on the interpretation of approximately 3600 boreholes for the area shown. The composition and thickness of made ground in this area are recorded in boreholes and this information is used to constrain geological models. In figure 8, made ground infilling the former course of the River Irwell and deposited adjacent to the Manchester Ship Canal is up to 10 m thick and comprises colliery spoil, furnace waste and ash probably deposited during construction of the canal between 1887 and 1894.
Artificial ground that is formed as a result of changing land use through time is commonly contaminated and is, therefore, a potential source of pollution. The land-use history of a site or the area in which it lies strongly influences not only the type of artificial ground, but also its chemical properties. In some cases, the anthropogenic chemical signatures of artificial ground, or the sediment into which the pollution may migrate, can be used as a marker for the source of contamination and its absolute age. Sediments recovered in the Mersey Estuary, NW England, show distinct contamination, with elevated concentrations of polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) and mercury [57,58]. These signatures, and their variation with depth, are interpreted to relate to the industrial development of the Mersey region, where elevated mercury concentrations are associated with the expansion of the chemicals industry on the banks of the estuary in the mid-nineteenth century. It is possible to relate these chemical signatures to the timing of the onset of industrial processes and, therefore, provide an indicative age of the estuarine sediments and the related artificial ground source onshore. In other cases, methods of relative dating of artificial ground can be derived from historical map and written data sources. Dearman et al.  illustrated the relative anthropogenic history of infilling and culverting in part of the catchment of the rivers Tyne and Wear, NE England. Relative dating and recognition of anthropogenic markers in the anthropogenic stratigraphical record have also been achieved using distinctive objects in the subsurface. Ager  highlights one approach during the investigation of old mining camps in the USA, using the changing shape of beer cans as an ‘evolutionary’ marker. Ager also refers to a French classification system of anthropogenic deposits using such markers into an Upper dustbinian/trashcanian (with plastic) and a Lower dustbinian/trashcanian (without plastic).
The conventional BGS approach to mapping and three-dimensional modelling of artificial ground does not explicitly describe its composition or attribute an age. However, lithological composition is an important factor in the characterization of human geological signature in the Anthropocene. The degree of lateral and vertical heterogeneity in artificial deposits in the subsurface is assumed to be high. However, the composition of made ground and engineered fill can be characterized based on its texture, grain size and lithological composition (whether reworked rock or soil or processed anthropogenic sediments). Both the grain size and composition of the material can be recorded. Schemes such as British Standards for the description of soils and rocks can be used to describe anthropogenic material . For material deposited as waste, a classification based on that described in the European Waste Catalogue might be applied to complement a description of the sediment .
The two-dimensional and three-dimensional classification and characterization of artificial ground could be further enhanced by applying an attribute that describes the relative or absolute age of artificial ground to derive the rates of anthropogenically driven land-use change.
5. Preservation potential of anthropogenic artificial ground
The classification and characterization of artificial ground provide a means to quantify the magnitude of anthropogenic activity. However, in geological time, this is dependent on its subsurface preservation. The natural geological record represents a complex legacy of processes. At one extreme, processes of accumulation result in the localized building of the rock record and the preservation of evidence of geological events and environments (e.g. sediments and fossils). At the other extreme, processes of erosion result in the removal and reworking of existing deposits and the introduction of time gaps (discontinuities) in the rock record. The result is an incomplete and often biased perspective on the geological history of an area. Processes responsible for artificial ground operate in a similar fashion: human activity includes the accumulation of material (typically, but not exclusively, in urban environments) and the widespread removal of both natural and existing artificial deposits. These processes operate alongside their natural equivalents. The preservation of artificial ground in the geological record is contingent on its ability to survive the erosive and transformative effects of both natural and human processes in terrestrial environments.
Artificial ground is predisposed for destruction. The most immediate threat to its preservation is that of changing land use in response to social and economic drivers, including the regeneration of existing urban and industrial development and the exploitation of material previously classified as waste (e.g. reworking abandoned mine waste tips to exploit historic ‘waste’ deposits [63,64] and landfill mining ). In the longer term, most artificial ground will succumb to natural processes as erosion seeks to re-establish equilibrium that has been exceeded by the creation of artificial ground. Once buried, anthropogenic material in the ground is subject to chemical degradation. For organic deposits, hydrogeological conditions surrounding the material and changes in those conditions related to land use and environmental change may determine their susceptibility to degradation . Fluctuations in water level control the redox potential of the subsurface environment, promoting the formation of corrosive chemical species. Studies in the heritage city of York, UK, have identified up to 8 m of waterlogged organic, anthropogenic deposits beneath the city. Dewatering and urban development have degraded the buried deposits in situ, reducing their preservation potential.
However, it should be noted that, when appropriate conditions prevail, the geological record has the potential to capture traces of fleeting events [66,67] and remarkably delicate lifeforms [68,69]. Traces of anthropogenic activity dated to 3649±109 BP comprising footprints and trackway infrastructure have been preserved on the coast of NW England , for example. The same may apply to the preservation of made ground and ‘habitation traces’ such as buildings, landfill sites and engineered structures .
Specific locales, such as river valleys, have always held an attraction for the settlement of human communities. Analysis of the British fluvial stratigraphic record indicates a wider variety of channel types and styles in the past than found today, following the widespread and large-scale land drainage and channelization that accompanied the Industrial Revolution . Low-energy river systems are accretional in nature and, coupled with high water tables between ca 5000–4000 BP and 2500 BP, which have influenced catchment hydrology, they have increased the likelihood of cultural deposit preservation . Conversely, in some areas, construction and development in river valleys have led to drainage of valley floors, which has resulted in the lowering of water tables and the oxidation and subsequent degradation of archaeological deposits . While the accumulation of fine-grained sediments can improve the chances of preservation, it can also mask the deposits so that they cannot be detected by traditional prospection techniques. River valley environments can only be effectively studied by a multi-disciplinary approach, which seeks to integrate archaeological and geomorphological evidence .
Although brick, concrete, steel, plastic and glass represent significant components of artificial deposits, their relative propensity for preservation varies considerably. A comprehensive account of the specific mechanisms affecting preservation potential is beyond the scope of this paper. In general, however, the longer-term preservation potential for material deposited in terrestrial environments is far lower than that for marine environments. This presents a particular challenge for the preservation of artificial ground that is preferentially developed on land. Natural terrestrial processes including the weathering effects of wind and ice and the erosive potential of relatively high-energy fluvial systems conspire to level the land surface (conveying material to the marine environment). The proportion of marine strata (including coastal delta deposits) in the 3 billion year old bedrock geological record in Great Britain is almost three times that for strata deposited in a terrestrial environment (Great Britain’s bedrock geology comprises approx. 57% marine deposits and 20% terrestrial deposits, with the balance of 23% represented by igneous or metamorphic strata).
It is, therefore, likely that artificial ground associated with coastal plain cities may have the greatest potential for preservation . In a scenario of rising sea level, the process of preservation may begin with abandonment (and the cessation of potentially destructive human processes) followed by inundation (and the abeyance of terrestrial weathering) and the eventual concealment beneath natural marine deposits, offset by the process of coastal erosion. With increased depth of burial, the effects of heat and pressure may commit the artificial ground to the sedimentary cycle and bedrock record.
An appropriate timeframe is essential when considering the preservation potential of anthropogenic activity, from predicting its preservation in the context of geological time (and the concept of the Anthropocene), to understanding the nature of evidence for early human activity or potential contamination associated with recent made ground. The anticipated response of anthropogenic deposits and landforms to each of these time ‘scales’ is an essential factor in understanding the legacy of human activity on the landscape.
Humans are a major factor in transforming the landscape, including the subsurface, through the deliberate shift of rock and soil, production of waste and enhanced denudation from agricultural activity. The physical transformation of the landscape correlates with increased population growth and the development of socioeconomic, political, technological and cultural drivers for landscape exploitation to meet their needs. The physical transformation can be measured in terms of its magnitude of impact (quantity of material moved) and its rate (the time over which the material movement occurred). Humans as geological and geomorphological agents are a major factor in British landscape evolution within the Anthropocene.
Rapid and large-scale anthropogenic activities are commonly associated with the Industrial Revolution. The sedimentological consequences of this are coincident with atmospheric release of gases such as CH4 and CO2 to the atmosphere. As such, the sedimentological response and its record in the subsurface can be considered as an anthropogenic marker that is broadly coincident with the Industrial Age of the Anthropocene proposed by Steffen et al. . The impacts of much earlier human activity as a landscape-transforming agent in the Anthropocene should not be discounted, however.
Anthropogenically driven landscape changes were already in place before the Industrial Revolution. The timing of the onset of humans as geological and geomorphological agents in Great Britain can be considered as that time in which earth-moving activities began. The transformation from hunter–gatherer to farming communities with small-scale subsurface exploitation for flint occurred at Grimes Graves during the Neolithic and is one example of early human activity as a landscape-transforming process through creation of artificial ground. The onset of Bronze Age resource exploitation, metal processing and agricultural activity on a larger geographical scale than previously seen represented a further transformation towards humans becoming significant factors in landscape evolution and a defining factor in the Anthropocene.
As with geological boundaries, the timing of onset of major anthropogenic activity is likely to have been different across different areas in Great Britain. The nature of this difference is likely to be even greater on a global scale. Therefore, the timing of significant anthropogenic activity as a geological and geomorphological process and its sedimentological impact is likely to be diachronous over tens, hundreds or even thousands of years when the developed and developing worlds are considered. The widespread exploitation of subsurface minerals, especially coal and iron, in Great Britain led to rapid transformation of the landscape as materials were excavated and waste deposited. Waste disposal, and large-scale urbanization, may represent the most significant visible record of widespread human activity in Great Britain. This record may also be preserved in the subsurface geological record. Future uses of the subsurface such as repositories for disposal of radioactive wastes may also provide a mechanism for the preservation of underground artificial ground.
The characterization of the drivers for anthropogenic processes, land-use change and the resulting artificial ground is essential to quantify the impacts of historical and current anthropogenic activity on the geosphere. The interaction of people and the shallow subsurface creates a shallow zone of human interaction where natural and anthropogenic processes take place. The sustainable management of this zone, and the ecosystem services that it provides to society, is critical if it is to function effectively in response to future environmental change. Its resilience to environmental change, including changes in land use, climate and population growth will determine its nature and properties within the Anthropocene Epoch.
The authors wish to acknowledge the assistance of colleagues and research partners, including Hans de Beer (Geological Survey of Norway), Jan Zalaseiwicz (University of Leicester), Martin Culshaw (University of Birmingham), Don Cameron, Jo Mankelow and Teresa Brown (BGS), Laurance Donnelley and Dave Wilshaw (Wardell Armstrong). The helpful review comments of Helen Reeves and Helen Burke at the BGS, Andy Howard (University of Birmingham) and Ian Douglas (University of Manchester) are also gratefully acknowledged. John Powell, Dave Lowe and Holger Kessler (BGS) are thanked for their contribution to the development of the classification schemes of artificial ground used in three-dimensional geological models and on maps published by the British Geological Survey. This article is published with the permission of the Executive Director of the British Geological Survey (NERC).
One contribution of 13 to a Theme Issue ‘The Anthropocene: a new epoch of geological time?’.
- This journal is © 2011 The Royal Society