Amazonian Dark Earths: Wim Sombroek's Vision

Amazonian dark earths are anthropogenic soils called terra preta de índio in Brazil, created by indigenous people hundreds, even thousands, of years ago (Smith 1980; Woods and McCann 1999). Terra preta proper is a black soil, associated with long-enduring Indian settlement sites and is filled with ceramics and other cultural debris. Brownish colored terra mulata, on the other hand, is much more extensive, generally surrounds the black midden soils, contains few artifacts, and apparently is the result of semi-intensive cultivation over long periods. Both forms are much more fertile than the surrounding highly weathered soils, mostly Ferralsols and Acrisols, and have generally sustained this fertility to the present despite the tropical climate and despite frequent or periodic cultivation. This fertility probably is because of high carbon content, which retains nutrients and moisture, and an associated high and persistent microbial activity.

The high concentrations of pyrogenic carbon in terra preta come mainly from charcoal from cooking and processing fires and settlement refuse burning, and in terra mulata the carbon probably comes from in-field burning of organic debris. Low intensity “cool” burning, what has been called slash-and-char, resulting in incomplete combustion, can produce carbon in high quantity which can persist in soil for thousands of years. Dated carbon in dark earths is as old as 450 BC (Hilbert 1968; Petersen et al. 2001:100). In contrast, slash and burn shifting cultivation fires today tend to be “hot” fires, set at the end of the dry season, which produce large releases of carbon dioxide to the atmosphere and more ash of brief persistence than charcoal.

During the past decade, integration of anthropology, archaeology, biology, ecology, geography, and soil science has brought important results in the development of an overview of the formation processes of Amazonian Dark Earths (ADE). These interdisciplinary efforts have provided significant information about the genesis, use, and re-use of these soils; moreover, this research is producing important information about pre-Columbian societies and the cultural behaviours that could start the ADE accretion in Amazonia (e.g. Lehmann et al. 2003; Glaser and Woods 2004; Rebellato 2007: Arroyo-Kalin, this volume; Schaan et al., this volume). An archaeological effort directed toward understanding the past socio-cultural processes responsible for the origin of these soils and the subsequent use is presented in this chapter. Evidence for continuity and change in settlement patterns during pre-Columbian times at the Hatahara archaeological site in the Central Amazon of Brazil is reviewed (Fig. 2.1). Soil analysis results correlated with archaeological artifacts excavated in that site provide interpretation of cultural changes, the consequences of these in village morphologies, and advance the interpretation of the region's indigenous history.

Located on a bluff top parallel to the left bank the Solimões/Amazon River, near to the confluence of the Negro and Solimões rivers, the Hatahara site has natural protection against attack due its 40 m high location. The scarp of this bluff surrounds ca. 60% of the site area leaving only a narrow entrance in the northeast. The secure location of the site was enhanced by access to a wide range of resources that allowed the population to survive by fishing, hunting, gathering, and farming (Fig. 2.2). The Hatahara site is an example of the Bluff Model described by Denevan (1996), who understood that pre-Columbian settlements in Amazonia were often located on bluffs adjacent to major river channels and their floodplains. These settlements, consequently, were not subject to the annual floods that cover the lowlands, but still had ready access to the fertile soils of the floodplain. Hatahara contains pottery from three different archaeological phases (the Açutuba — c.300 BC—c. AD 360; Manacapuru c. AD 400—800; Paredäo c. AD 700–1200; and, the Guarita Subtradition c. AD 900–1600) (Heckenberger et al. 1999; Hilbert 1968).

In this chapter analyses the results from soils sampled in the Hatahara archaeological site. The first objective of this proposal was to determine if the Hatahara had an agricultural field during the pre-Columbian times and, if validated the agricultural use of the land, to identify the plants that were cultivated in the past. The general site characteristics and location were presented in earlier chapters (see Rebellato et al. Chapter 2 in this volume; Arroyo-Kalin Chapter 3 in this volume). The samples were collected from a ca. 1.5 m column through one of the site's mounds (M-I; see Fig. 2.2 this volume). The mound is associated with the Paredao phase occupation of the site. The mound was constructed with dark earth soil from the surrounding area that had developed during the previous Manacapuru occupation (see Machado 2005). Consequently, the archaeological materials, including the phytoliths, found within it are not in situ and are from the disturbed context developed by this earthmoving operation ca. 1,000 years ago. Even the terra preta below the mound is not in primary context. It was disturbed by a burial program that occurred during Manacapuru phase c500 years before the construction of the mound. A total of 22 skeletons was recovered from the submound context. So, there are basically two contexts represented by these samples: (1) The redeposited materials from within the mound; and (2) The turbated materials found below the mound. The latter are at least close to their original place of deposition and were sealed by the mound. Within the mound the materials are most likely in inverse stratigraphic/temporal position and come from a much wider area. In spite of the disturbed context for these samples it is felt that the study was available one in that it demonstrated the array of useful plants found in association with the occupations of the Hatahara site. As will be seen the phytolith data derived from this study (one of the very few so far in Amazonia) are in conformance with the stance of that a broadlybased subsistence strategy was operant in the Central Amazon during the pre Columbian period.

From this vantage point, with a sincere hope of contributing to the broader enterprise of understanding past anthropogenic landscape transformations in the Amazon basin, and in fond memory of Wim Sombroek, undoubtedly a pioneer in investigations of these soils, we offer below two interlinked sets of remarks about the physical and chemical characteristics of anthropogenic dark earths. We focus first on dimensions of the variability and evolution of the soil mantle of the central Amazon region that clearly impinge on the physical and chemical properties of these soils and, therefore, affect the nature of archaeological inferences that can be drawn from them.

A great deal of Amazonian research has been devoted to the study of indigenous management of tropical forest resources. Questions of soils research have become particularly significant for understanding the forms of landscape manipulations that could have given rise to large pre-Columbian populations in tropical forest areas, and their implications for sustainable practices today, especially in light of deter-minist theories that viewed the possibilities of Amazonian cultural evolution as constrained by the general poverty of its upland soils (Denevan 2001; Erickson 2000; Heckenberger et al. 2003; Kim et al. 2007; Lehmann 2003; Peterson et al. 2001).

The analyses of indigenous tropical savanna management, and especially of soils on the other hand, are relatively rare. In part this reflects the reality that “savanna” peoples are also often managers of forests, since savanna ecosystems are usually part of ecotones that include gallery forests and a complex mosaic of lower biomass, semi-deciduous forests and successional systems as well as the open woodlands. Researchers often laid the emphasis on the forest rather than the Cerrado side of the equation, because they themselves were often more interested in the agricultural bases of these societies (Gross, Hames). For funding, it was often convenient to portray such “heterodox” groups as forest dwellers, or ambiguously as “native Amazonians” since such an approach fit in with broader social and scientific concerns about tropical environments and human rights alarms over Amazonian development.

The Kuikuro Amerindian word for the black soil is igepe, which is also their word for cornfield. They plant fields of their staple crop manioc on the normal red soils but all other crops are planted on patches of deep terra preta with abundant ceramic fragments and boasting large prehistoric earthworks that give structure to the sites. They plant banana, corn, squash, sweet potato, papaya, sugarcane, and many other crops that generally would not do well in the natural red soil. The fact that they are willing to travel up to 10 km or more to reach these igepe gardens and haul the produce back to the village demonstrates the value of the terra preta to the Kuikuro. They are using a resource that was accumulated over decades or centuries by previous inhabitants but, are they also creating it themselves?

Throughout the world intensive agricultural land use often has resulted in soil physical and chemical degradation, due to erosion and higher output than input rates of nutrients and OM. In contrast, the intentional and unintentional deposition of nutrient-rich materials within human habitation sites and field areas has in many cases produced conditions of heightened fertility status (Woods 2003). An anthropogenic enriched dark soil found throughout the lowland portion of the Amazon Basin and termed Amazonian Dark Earths (ADE) or terra preta de índio (TP) is one such example. Its fertility is the secondary result of the transport of natural and produced foods, building materials, and fuel to prehistoric dwelling places (Woods 1995). These materials and their byproducts were then transformed and differentially distributed within the zone of habitation and associated garden areas. The resulting soil contains high concentrations of black carbon (C) as charcoal (Glaser et al. 2001a); significantly more C, nitrogen (N), calcium (Ca), potassium (K), and up to 13.9 g kg⁻¹ P2O (almost 4 g kg⁻¹ available P) (Lima et al. 2002); and significantly higher cation exchange capacity (CEC), base saturation (BS), and pH values than in the surrounding Oxisols (Glaser et al. 2000; Zech et al. 1990). ADE is found at pre-Columbian settlements throughout Amazonia in patches ranging in size from less than a hectare to many square kilometres (Woods and McCann 1999). Today these soils are and presumably in the past were intensively cultivated by the rural population. The enormous total area encompassed by these soils suggests a large sedentary pre-Columbian population (Erickson 2000; Heckenberger et al. 2003). The existence of ADE proves that infertile Ferralsols and Acrisols can be transformed into permanently fertile and refractory TP in spite of rates of weathering 100 times greater than those found in the mid-latitudes. Such a transformation cannot be achieved solely by replenishing the mineral nutrient supply, however; the soil organic matter (SOM) is also of prime importance for insuring the retention of soil nutrients (Zech et al. 1990).

A great challenge for the world's scientific community has been to find alternatives for the best use, management and conservation of the Amazonian tropical rainforest. The unbalance between increasing populations and the demand for food shows the urgency for looking for new alternatives that region, avoiding excessive degradation of natural ecosystems. In this respect, scientists in all countries in the tropics are working to understand the functioning of the rainforest and come up with alternatives to use and manage this natural resource, applying the best technology not only from an economic perspective, but principally from social and ecological perspectives. The replacement of the nutrients exported by crops is normally solved by inputs of chemical and organic fertilizers, in proportions that can vary with natural soil fertility and the nature and volume of the crops. Hence, problems like erosion, leaching, and compactation can only be solved by conservation practices utilized by farmers that change as function of land topography, precipitation, crop type, cover crop, or farming land use system, depending on the technology level of the community.

Small farmers living in Manacapuru, Iranduba, Presidente Figueiredo, and Rio Preto da Eva, in Amazonas, Brazil, work on Amazonian Dark Earth (ADE) sites, cultivating vegetables and perennial crops like oranges, coconuts, cupuaçu, and others. Their land use systems include monocultures, two species mixes, and agrofor-estry systems. In general, the small farmers report that ADE is very fertile and they never need to apply chemical and organic fertilizers to get high productivity. On the other hand, studies have shown that ADE presents some nutrient limitations to plant production. Some small farmers are using large amounts of chemical and organic fertilizers and even liming unnecessarily. This intensive exploitation and the excessive use of nutrients is causing chemical degradation (Falcão et al. 2003) and even physical degradation of ADE (Teixeira and Martins 2003).

This chapter explores how the recent historical ecology of two distinct riverine settings in Central Amazonia has shaped the contemporary use (and non-use) of Amazonian Dark Earths1 (ADE). Drawing on ethnography and oral histories from ongoing fieldwork on the Middle Madeira and Lower Negro rivers it asserts that patterns of land use on ADE by the Amazonian peasantries and Indigenous groups resident in each region are conditioned by divergent agro-ecological histories. The correspondence of life histories and particular configurations of agro-extractivist activity amongst informants suggests wider trends in the relation of livelihood to ADE. The distinctive processes of kinship and sedentary or mobile settlement encountered in each region are shown to be contingent on different patterns of social and economic history within and between the two great rivers. Kinship ties, permanency of residence and social networks are identified as being critical factors in enabling the practice of agriculture. The chapter reconstructs the divergent agro-ecological trajectories that have engendered the widespread practice of agriculture on the Middle Madeira, and the prevalence of extractivism in livelihood trajectories on the Lower Negro, until recent (1950 and onwards) migrants from the more agricultural regions of the Solimões and Upper Negro Rivers have begun to forge a local agricultural tradition. It argues that these current patterns of land-use — rather than being ‘adaptive’ strategies in response to certain static environmental constraints (Moran 1993) — have emerged vis-à-vis landscapes shaped by the interaction of historical and environmental factors over time (Baleé and Erickson 2006).

The historical ecology of the Lower Negro is shown to have constrained agriculture. People successfully farming ADE there are those who have migrated from regions where agriculture is more prevalent. In contrast, the historical ecology of the Middle Madeira is shown to have enabled agriculture. This implies that the longer-term trajectory of agriculture on the Middle Madeira (and for those families on the Negro from agricultural backgrounds) should have allowed the evolution of a greater repertoire of local knowledge related to ADE. The cultivation of ADE and awareness of its anthropogenic origins are examples of how local knowledge and ethnobiological perceptions are contingent on historical ecology. The chapter demonstrates the utility of comparative historical ecology in understanding contemporary agriculture on ADE in different regions. It concludes that further comparative research is important in order to explore the significance of manioc (Manihot escu- lenta) agriculture on ADE and how local understandings of the cultivation, management and creation of ADE could be incorporated into the creation of Terra Preta Nova.

During the last 20 years, research on Amazonian soils has been central to a complete reappraisal of the region's social and natural history. Patches of dark and highly fertile soils have been found to occur throughout Amazonia, known as Amazonian Dark Earths (ADE) and sometimes distinguished as terra preta (Black Earths) and terra mulata (Brown Earths). The former are usually described as the legacy of the former settlement sites (middens) of pre-Colombian farmers, and the latter as a legacy of their agricultural practices. The ability of these soils to support intensive agriculture has undermined environmentally-determinist views of Amazonian history which until recently asserted that the inherently infertile soils could not support populous settled farming.

The importance of ADE is not restricted to their historical significance. First, these soils are sought after by today's farmers (Woods and McCann 1999; German 2003; Fraser et al., this volume). Second, the development of new techniques to establish them rapidly could help intensify modern farming in Amazonia and beyond. Third, because the secret to these soils is at least partially due to the high proportion of charred carbon they contain, farming technologies based on ADE have the potential to sequester enormous quantities of carbon, suggesting a ‘win-win’ opportunity, improving sustainable agriculture whilst mitigating climate change.

ADE sites vary in their size and degree of engineering (Woods and McCann 1999), and their biochemistry may be influenced by the addition of any of the following: large amounts of pottery sherds, concentrated organic wastes, charred biomass, fish bones, shells, various household wastes. In many ADE these additions have resulted in notably high nutrient concentrations of calcium, phosphorus, and potassium, and also high levels of black carbon (BC). The latter is thought to play a key role in greater nutrient retention and stabilization of soil organic matter, as decomposing residue or within living cells (Glaser et al. 2003; Sombroek et al. 2003), in spite of intense weathering conditions which typically lead to highly leached soils in the humid tropics (Lehmann et al. 2003). Thus the central role of the soil microbial community is in this unique soil environment should be traced to soil amendments such as BC. Black C is not unique to ADE, but occurs throughout terrestrial and aquatic environments (Schmidt and Noack 2000) as residue from naturally-occurring and human-induced burning. Once created, BC persists over time-scales of millennia and is thought be highly recalcitrant to microbial degradation (Schmidt and Noack 2000). In soil, BC may enhance soil fertility by decreasing bulk density, improving moisture retention and increasing pH, and the surface charge properties of BC are thought to increase cation exchange capacity (CEC), thereby reducing nutrient leaching (Glaser et al. 2002). The anthropic addition of BC to ADE may help stabilize inorganic nutrients and thus maintain soil fertility (Glaser et al. 2003), however, it may also serve directly as a habitat or as a platform for nutrient exchange for microorganisms (Abu-Salah et al. 1996 Chitra et al. 1996). High BC additions to soil due to fire events in northern boreal forest, have been shown to alter soil microbiological community structure and above ground plant communities (Pietikainen et al. 2000; Zackrisson et al. 1996), but little other research exists on the impact of BC on below-ground biological communities. Due to its prevalence in ADE and its unique physical and chemical characteristics, BC is thought to be central to the biogeochemistry of these soils.

Charcoal is a major component of stable SOM in terras pretas alsocalled Amazonian Dark Earths (Glaser et al. 2001a, b; Glaser 2007). Apart from charcoal, special microbes could contribute to the formation of the highlystable SOM in terra preta (Woods and McCann 1999). However, this is stillmatter of speculation. There could be a link between the high amounts of charcoal in terra preta soils and soil microbial community composition (Glaser 2007). Although it is unlikely that charcoal is used by microorganisms as a direct carbon source, habitat properties are certainly different with the presence or absenceof charcoal (Saito and Marumoto 2002). Steiner et al. (2004) could demonstrate that charcoal addition to Ferralsols significantly increased microbial activity. Charcoal also promoted the colonization of agricultural plants with arbuscular mycorrhizal fungi (Nishio 1996; Saito and Marumoto 2002), it improved nodule weight (Nishio 1996) and nitrogen fixation (Tryon 1948; Nishio 1996; Rondon et al. 2007).

Most soil microorganisms cannot be characterized by conventional cultivation techniques (Zelles 1999). Amann et al. (1995) estimated that 80–99% of all micro-bial species have not yet been cultured. Therefore, biomarkers such as ribosomal nucleic acids (RNA) or phospholipid fatty acids (PLFA) are better analytical tools to provide an unbiased view on the structure of complex soil microbial communities (Zelles 1999). PLFA are exclusively found in the membranes of living organisms and comprise a relatively constant portion of living biomass within a microbial community (Zelles 1999). The analysis of PLFA allows the simultaneous quantification of microbial biomass and the characterization of the microbial community structure by specific biomarker PLFA (Tunlid and White 1992; Zelles 1999; Gattinger 2001).

P is usually considered the primary limiting nutrient in plant production on highly weathered soils of the humid tropics because it is strongly bound to aluminium and iron oxides and, thus, not easily available for plants (Garcia-Montiel et al. 2000). Heterophobic phosphate solubilizing microorganisms make mineral bound P available by the excretion of chelating organic acids. (Kimura and Nishio 1989) showed that insoluble phosphates which are not crystallized can be solubi-lized by indigenous microorganisms when abundant carbon sources are supplied.

In the Amazon soils occur that were formed during pre-historical human occupations. These soils are highly fertile and stable and are most commonly referred to as terra preta, Amazonian Dark Earths (ADE) (Woods and McCann 1999), or Archaeological Black Earth or Indigenous Black Earth (Kern 1988, 1996; Kern and Kämpf 1989). It appears that ADE form proper micro-ecosystems where the soils do not exhaust themselves easily, even in the tropical conditions where are exposed for long periods of time. Several process-oriented designations have been suggested for the ADE formation processes, such as: “vegetable soils”, “plaggen epipedon” or “anthropic soils”. The latter one, the most accepted, proposes that the ADE were intentionally formed by the pre-historical inhabitants. Archaeological evidence indicates that ancient human activities in the Amazonian habitats transformed, significantly, the landscapes of the vicinity of their settlements (e.g. Kämpf and Kern 2005). In many areas, indigenous societies formed extensive deposits that altered the soil properties (Lehmann et al. 2003; Woods and McCann 1999). These archaeological sites' dark soils were created by deposits of vegetal origin (charcoal, ash, leaves and diverse palm fronds, manioc residue, seeds, etc.) and residues of animal origin (bones, blood, fat, feces, chelonian carapaces, shells, etc.). These materials resulted in highly fertile soils with elevated levels of P (more than 1,000 mg/kg⁻¹ of soil), Ca, Mg, Zn, Mn, and C (Kern 1996). The organic matter in ADE is on the order of six times more stable than in the soils of forest (Pabst 1992), owing to the stability of fertility in ADE. In locations where Black Earth is present a strip of soil with dark brown coloration is often also found, without ceramics artifacts, but with elevated levels of OM. These soils were called terra mulata (TM) by Sombroek (1966). According to Sombroek et al. 2002, the areas of TM are the resultant of the intentional application of charred plant materials sometimes associated with human or animal residues (fishing and hunting products, calcium from shells and mollusks from consumption) and from other residues of plant origin. The development of the terra mulata expanses associated with the habitation areas allowed for expansion of horticultural food production and most reliability of sustainable harvests.

Many aspects of the origin of Amazonian Dark Earths (ADE) are still unclear. The analysis of Amazonian anthropogenic soils indicates that the alterations caused by human actions, such as the incorporation of organic residues and the effects of fire on the upper horizons influenced many of the chemical and physical characteristics of these soils (e.g. Woods 1995; Kern 1996; Glaser 1999; Woods and McCann 1999; Ruivo and Cunha 2003; Ruivo et al. 2004). Ruivo et al. (2004) show that the microbial biodiversity is higher in ADEs than in Yellow Latosols. ADE soils are more aggregated than Yellow Latosols, a factor that facilitates soil aeration, root distribution, water retention and movement.

Soil is an important habitat for microorganisms. The biology of ADE soils is important. Little is currently known about the abundance, activity, and diversity of organisms extant in ADE. So, we have much to learn (e.g. Thies and Suzuki 2003; Tsai et al. this volume, Chapter 15). The study of local knowledge of soil and land management in an ecological perspective in the Amazon Region is very important. The soil microbiological relationships are important for soil fertility management. Analysis of the past and present can help make recommendations on how ethnopedological studies can contribute to enhanced sustainable land use and management in the Amazon (Lehmann et al. 2003; WinklerPrins and Barrera-Bassols 2004; Silva and Rebellato 2004; Thies and Suzuki 2003).

Soils are one of the most important natural resources, and are essential for the development and continuation of any society that practices agriculture. Ancient civilizations in the Old World generally began in valley regions and floodplains along big rivers. As examples, one can cite Egypt in the Nile River valley, the Mesopotamia between the Tigris and Euphrates rivers, the Indian Subcontinent by the margins of the Indus and Ganges rivers, and China in the valleys of the Yellow and Blue rivers. In different areas of the world, many detailed comparisons have been accomplished between natural and human influenced soils, and the results of the latter have been documented (Cunha 2005)

In the last decades, agricultural activities have been modifying the original vegetation cover of a great part of the Brazilian territory. Ecosystems, such as the Amazon forest, are losing their original characteristics, being replaced by agricultural and extractive activities. The expansion of the agricultural borders has been causing great changes in forest areas in Brazil, with the introduction of rice, soybeans, and pastures, mainly in the southern part of the Amazon area. This has lead to degradation and a loss of biodiversity, a reduction in organic matter, and also the degradation of the pedologic covering, through the exhaustion and erosion.

The environmental organic matter is the link between the biosphere, geosphere, hydrosphere and atmosphere and is fundamental for ecosystem sustainability. Estimates of the total mass of organic carbon in soils are in the range of 1.22 × 10¹⁸g (Sombroek et al. 1993) to 2.456 × 10¹⁸ g (Batjes 1996). This reservoir is at least three times greater than all organic materials above the earth's surface, and the way the soil sequestered carbon is managed can have significant influences the levels of atmospheric CO₂. Estimates of the amounts of fossil organic carbon (gas, oil, coal etc.) are considerably greater, of the order of 4 × 10¹⁸ g (Falkowski et al. 2000; Janzen 2004). An increase of carbon improves the fertility of soil, especially in tropical conditions, and thus increases the vegetal biomass that this soil can support.

The soil carbon stock represents a continuous process of deposition (5.67 × 10¹⁶g C year⁻¹), in the form of vegetal and animal residues, and decomposition (with emissions of 5.50 × 10¹⁶ g C year⁻¹). The deposition and decomposition fluxes are not equal because of the inputs of fossil fuel carbon (5 × 10¹⁵ g C year⁻¹). However, fossil fuel carbon emissions are one order of magnitude less than that due to the decomposition of natural soil organic matter (United States 1999).

The soils of the Amazon region are usually unfertile, due to the high decomposition rate of organic carbon (C), rapid losses of nitrogen (N) and potassium (K) through leaching, and rapid phosphorus (P) fixation to (hydr-) oxides of iron (Fe) and aluminium (Al). Consequently, these soils (oxisols, ultisols) are known for their low suitability for agricultural production purposes. However, relatively fertile, pH-neutral soils also occur in the Amazonian lowlands. Such soils of high fertility are remnants of ancient pre-Columbian inhabitants. They are known as terra preta de índio or Amazonian Dark Earths (ADE) (Woods and McCann 1999). Terra preta contain up to eight times more carbon than adjacent soils. Furthermore, available and total nitrogen is two to eight times higher and there is up to 1,000 times more available phosphorus and up to ten times more total phosphorus than in adjacent soils (Lehmann et al. 2003).

Five years after Wim Sombroek, who can be considered the founder of research on terra mulata, passed away, his legacy is still alive in Wageningen, the city where he spent much of his career. Anthrosols are common in many areas in the Netherlands and, consequently, much research has been done on these soils in Wageningen. Sombroek's parents used to farm so called eerdgronden (Plaggen soils). The soil formation history of ADE shows similarities to these eerdgronden and both types are classified into the FAO-category of man-made anthrosols. To compare, P levels of eerdgronden (currently under forest) vary from 375 to 3,000 mg kg⁻¹ (TW Kuyper, unpublished data, 2008) and ADE from 73 to 8,800 mg kg⁻¹ (e.g. Costa and Kern 1999; Glaser 1999; Madari et al. 2003; Ruivo et al. 2003; Lehmann et al. 2003; Schaefer et al. 2004), whereas adjacent soils (non man-made) contain on average 150–300 mg kg⁻¹ (Guttmann et al. 2006). C levels in Northern European sandy anthrosols vary from 1.3% to 2.8% (Blume and Leinweber 2004) which is lower than C levels in ADE which are between 2.8% and 9.0% (Costa and Kern 1999; Glaser 1999; Lehmann et al. 2003; Madari et al. 2003).

Charcoal production worldwide is increasing for energy use in households and industry, but it is often regarded as an unsustainable practice and is linked to agricultural frontiers (Prado 2000). The production (Coomes and Burt 1999) and use of charcoal in agriculture is common in Brazil and widespread in Asia (Steiner et al. 2004).

The efficiency of biomass conversion into charcoal becomes important in conjunction with a newly proposed opportunity to use charcoal as a soil conditioner that improves soil quality on very acid and highly weathered soils (Lehmann et al. 2002; Steiner et al. 2004). This can be realized either by charring the entire above-ground woody biomass in a shifting cultivation system as an alternative to slash-and-burn (coined recently as slash-and-char by (Glaser et al. 2002; Lehmann et al. 2002) or by utilizing crop residues in permanent cropping systems. Charcoal formation during biomass burning is considered one of the few ways that C is transferred to refractory long-term pools (Glaser et al. 2001a; Kuhlbusch and Crutzen 1995; Skjemstad 2001). Producing charcoal for soil amelioration instead of burning biomass would result in increased refractory soil organic matter, greater soil fertility and a sink of CO₂ if re-growing vegetation (secondary forest) is used. A farmer practicing slash and char could profit from soil fertility improvement and C credits (if provided by a C trade mechanism to mitigate climate change), providing a strong incentive to avoid deforestation of remaining primary tropical forests.

Agricultural production in the tropics is frequently limited by low soil fertility. Many tropical soils are rich in kaolinite and iron (Fe) and aluminium (Al) oxides, but relatively poor in soil organic matter (SOM), they frequently have a low cation exchange capacity (CEC), low pH, low calcium (Ca), and phosphorus (P) contents (Zech et al. 1990). In strongly weathered tropical soils, SOM plays a major role in soil productivity because it represents the dominant reservoir of plant nutrients (Tiessen et al. 1994; Zech et al. 1997). Generally SOM contains 95% or more of the nitrogen (N) and sulphur (S), and between 20% and 75% of the P in surface soils (Duxbury et al. 1989). Rapid mineralization of organic matter after clearing of forests and during continuous farming explains low fertility levels of many tropical soils under permanent cropping systems (Tiessen et al. 1994; Zech et al. 1990).

Conventional fertilization with mineral fertilizer (mainly N applied as urea or ammonium sulphate) is not very efficient on soils with low nutrient retention capacity (Alfaia et al. 2000). Heavy tropical rains leach easily available and mobile nutrients into the subsoil where they are unavailable for most crop plants (Giardina et al. 2000; Hölscher et al. 1997; Renck and Lehmann 2004). To overcome the poor nutrient supply the common agricultural practice in the tropics is slash and burn agriculture. Practiced by about 300 to 500 million people worldwide, (Giardina et al. 2000; Goldammer 1993) slash and burn contributes significantly to global warming (Fearnside 1997). This traditional agricultural practice is considered sustainable if adequate (up to 20 years) fallow periods follow a short time of cultivation (Kleinman et al. 1995). Long fallow periods make this technique land demanding and hardly any other crop than manioc can be cultivated by shifting cultivation in Central Amazon, without access to fertilizers.

Char produced from the pyrolysis of biomass has potential as an agricultural amendment in low fertility soils. Much of the interest in its use as an agricultural amendment has been stimulated by research discussed in this book and the previous volumes on the role of charcoal in terra preta soils. Results from studies conducted in South American and African tropics on acidic, highly-weathered Oxisols with low organic carbon (C), cation exchange capacity (CEC), and base saturation indicate that addition of charcoal has significantly influenced nutrient cycling, soil biology, and crop productivity (Glaser et al. 2002; Lehmann and Rondon 2006; Oguntunde et al. 2004). Increased yields and biomass have been reported for various legumes (Iswaran et al. 1980; Lehmann et al. 2003; Topoliantz et al. 2005) and for corn (Lehmann and Rondon 2006; Oguntunde et al. 2004). Increased productivity may be related to available nutrients (Glaser et al. 2002; Lehmann et al. 2003; Steiner et al. 2007), or increases in pH (Topoliantz et al. 2005; Steiner et al. 2007), and CEC (Steiner et al. 2007; Liang et al. 2006), as well as changes in water relations and soil biology (Glaser et al. 2002; Steiner et al. 2004). Although most studies report increased plant productivity with charcoal addition, plant biomass decreases have been observed, particularly at high application rates (Glaser et al. 2002). These responses could be related to nitrogen immobilization through high C:N ratios and sorption of NH₄ and NO₃ (Lehmann and Rondon 2006).

The southeastern United States is an important agricultural area. The state of Georgia alone has approximately 4.3 million hectares of corn (Zea mays), soybean (Glycine max), cotton (Gossypium hirsutum), and peanuts (Arachis hypogaea) in production and 9.6 million hectares of forestland largely in loblolly pine (Pinus taeda) production (USDA 2002; Georgia Forestry Commission 2007). The growing interest in biofuels is increasing demands on row crop production and may also increase demand on forestlands. The Ultisols of the southeastern United States are similar to tropical Oxisols with low organic C concentrations of less than 1%, low CECs of approximately 5 cmol kg⁻¹, and low base saturation of usually less than 30% (Perkins 1987). Char from energy production through pyrolysis may provide an opportunity to increase the productivity of southeastern soils, similar to the way charcoal functions in terra preta. However, because char characteristics vary with feedstock and pyrolysis conditions (Harris et al. 2006; Antal and Gronli 2003), a better understanding of the influence of these factors on char characteristics and the effect of different chars on soil processes in the southeastern United States is needed.

The total amount of crop residues produced each year in rice-based systems of Asia can be roughly estimated at about 560 million tons of rice straw and about 112 million tons of rice husks (based on 2005 production, a harvest index of 0.5, and a husk/paddy ratio of 0.2). These residues constitute a valuable resource, but actual residue management practices do not use their potential adequately and often cause negative environmental consequences. In the past decades, increasing opportunity costs of organic fertilizer use and shortened fallow periods due to cropping intensification caused a continuous decline in the recycling of crop residues (Pandey 1998). Residue burning is widely practiced and causes air pollution, human health problems, and considerable nutrient losses. The declining return of organic materials to soils does not seem to affect soil quality in mostly anaerobic systems (rice-rice) with good soils but residue recycling is important to maintain soil fertility on poor lowland soils, in mixed cropping systems (rice-upland crop), and in upland systems (Dawe et al. 2003; Ladha et al. 2003; Tirol-Padre and Ladha 2006). Global climate change raises further questions about rice residue management. Decomposition of organic matter in flooded rice is always related to emissions of methane, which is about 22 times more radiatively active than CO2, and rice-based systems are estimated to contribute 9% to 19% of global methane emissions (Denman et al. 2007). In addition, the rapidly increasing interest in renewable energy sources adds new options and consequences for rice residue management and rice-based systems.

The current politically based model of economics seldom recognises all theelements of an economy and indeed has ignored to the peril and destructionof many communities that the basis of any economy is its soil. Any communityis literally built from the ground up. It is the quality of the soils of all nationswhich determine their viability, both in economic and social terms. Communityis built on agriculture. It always has been. Agriculture, in turn, is responsiblefor up to 70% of the industrial inputs of some economies (Chino 2001). If you have no soil you have no industry, if you have no soil, you have no community.

The basis of the relationship between a community and its soil — a community and its farmers, will be the relationship which determines the future stability and survival of humanity. The simple act of returning organic waste to agricultural soil has long been recognised as a logical and fundamental practice in many settled communities around the world. Indeed it is this very practice which contributed to the development of all the successful agricultural models of human history (Diamond 1999).

However, the lessons that ADE can teach us do not hinge upon the fact whether or not the Amazonian populations intentionally created these fertile soils for improving soil productivity for agriculture or whether they are an accidental byproduct of habitation. We can even draw the most important conclusions without ever knowing how ADE was actually created. These lessons can be gleaned from the properties of ADE today and the fact that these were in some way ‘created’ at a particular point in history a long time ago. As we can understand it today, the most important aspect of ADE is its high nutrient availability and high organic matter content. The goal of the recent efforts in ADE research has therefore been to find the answer to the question how it is possible that these favorable properties can still be observed after such a long period of time. What is unique about ADE that explains its sustainable productivity? Some of these lessons will be discussed in the first part of this chapter. In the second part, one of these lessons will be discussed in more detail with respect to the development of a new soil and biomass management approach: biochar agriculture for environmental management.