Evidence confirms an anthropic origin of Amazonian Dark Earths

MATTERS ARISING https://doi.org/10.1038/s41467-022-31064-2 OPEN 1234567890():,; Evidence confirms an anthropic origin of Amazonian Dark Earths Umberto Lombardo 1,2 ✉, Manuel Arroyo-Kalin 3 ✉, Morgan Schmidt 4, Hans Huisman5, Helena P. Lima 6, Claide de Paula Moraes 7, Eduardo G. Neves8, Charles R. Clement9, João Aires da Fonseca10, Fernando Ozorio de Almeida 11, Carlos Francisco Brazão Vieira Alho 12, Christopher Bronk Ramsey 13, George G. Brown 14, Marta S. Cavallini8, Marcondes Lima da Costa 15, Luís Cunha 16, Lúcia Helena C. dos Anjos17, William M. Denevan18, Carlos Fausto19,20, Caroline Fernandes Caromano 21, Ademir Fontana22, Bruna Franchetto19, Bruno Glaser23, Michael J. Heckenberger24, Susanna Hecht25,26, Vinicius Honorato7, Klaus A. Jarosch 2, André Braga Junqueira 1, Thiago Kater 8, Eduardo K. Tamanaha27, Thomas W. Kuyper 12, Johannes Lehmann 28, Marco Madella 29,30, S. Yoshi Maezumi 31,32, Leandro Matthews Cascon33, Francis E. Mayle 34, Doyle McKey 35, Bruno Moraes 36, Gaspar Morcote-Ríos37, Carlos A. Palheta Barbosa38, Marcos Pereira Magalhães 6, Gabriela Prestes-Carneiro7, Francisco Pugliese8, Fabiano N. Pupim39, Marco F. Raczka 34, Anne Rapp Py-Daniel7, Philip Riris40, Bruna Cigaran da Rocha 7, Leonor Rodrigues41, Stéphen Rostain42, Rodrigo Santana Macedo43, Myrtle P. Shock 7, Tobias Sprafke2,44, Filippo Stampanoni Bassi45, Raoni Valle7, Pablo Vidal-Torrado 46, Ximena S. Villagrán8, Jennifer Watling 8, Sadie L. Weber8 & Wenceslau Geraldes Teixeira 22 ARISING FROM Silva et al. Nature Communications https://doi.org/10.1038/s41467-020-20184-2 (2021) irst described over 120 years ago in Brazil, Amazonian Dark Earths (ADEs) are expanses of dark soil that are exceptionally fertile and contain large quantities of archaeological artefacts. The elevated fertility of the dark and often deep A horizon of ADEs is widely regarded as an outcome of preColumbian human influence1. Archaeological research provides clear evidence that their widespread formation in lowland South America was concentrated in the Late Holocene, an outcome of sharp human population growth that peaked towards 1000 BP2–4. In their recent paper Silva et al.5 argue that the higher fertility of ADEs is principally a result of fluvial deposition and, as a corollary, that pre-Columbian peoples just made use of these locales, contributing little to their enhanced nutrient status. Soil formation is inherently complex and often difficult to interpret, requiring a combination of geochemical data, stratigraphy, and dating. Although Silva et al. use this combination of methods to make their case5, their hypothesis, based on the analysis of a single ADE site and its immediate surroundings F (Caldeirão, see maps in Silva et al.5), is too limited to distinguish among the multiple possible mechanisms for ADE formation. Moreover, it disregards or misreads a wealth of evidence produced by archaeologists, soil scientists, geographers and anthropologists, showing that ADEs are anthropic soils formed on land surfaces enriched by inputs associated with pre-Columbian sedentary settlement6–9. To be accepted, and be pertinent at a regional level, Silva et al.’s hypothesis5 would need to be supported by solid evidence (from numerous ADE sites), which we demonstrate is lacking. Geomorphological and pedological considerations There are several problems with reviving the argument10 that ADE fertility originates from deposited alluvium. First, the Caldeirão ADE site is located on a Miocene plateau ~20 m above the Solimões River floodplain (~40 m asl), which in itself precludes significant flooding during the Holocene11. Second, A full list of author affiliations appears at the end of the paper. NATURE COMMUNICATIONS | (2022)13:3444 | https://doi.org/10.1038/s41467-022-31064-2 | www.nature.com/naturecommunications 1 MATTERS ARISING NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-022-31064-2 Fig. 1 Caldeirão’s soil compositional data compared with published data of Solimões River sediments and anthropic materials. Data is in Supplementary Table 1; the wood ash and bone/dung fields in (C, D) are offset to compensate for soil (Ulti) background concentrations. ADE = Amazonian Dark Earth, Ulti = Ultisol soil profile. A Geogenic elements Al and Fe are similar in ADE and Ultisols, but different from Solimoes sediments. B, C Anthropogenic elements K, Ca, and P fall in the range of anthropogenic materials. Solimões sediments have much lower Ca/K ratios and far higher K concentrations. Black continuous and broken lines give the 1:2 and 1:2.13 Ca:P ratios quoted by Silva et al.5 for human faeces and freshwater fish, respectively, corrected for 500 mg/kg soil (Ulti) background. D Ca and Sr show strong correlations in ADE. The Ca/Sr ratio in ADE is close to that of wood ash, suggesting an anthropogenic origin for Sr, while Solimões sediments have overall much higher values. the parent material of the ADE and adjacent Ultisol shows analogous clay mineralogy and geogenic composition: both sites are characterised by the same 1:1 clays (as shown by Silva et al.’s Supplementary Fig. 35) and both lack the 2:1 clay minerals expected from fluvial origin12. Moreover, no difference is observed in the geogenic elements (Al, Ti, Cr, V, Fe, As) (Fig. 1A). Third, the overall mineral assemblage of the Caldeirão ADE is incompatible with the geochemistry of the sedimentary load of the Solimões River (Fig. 1A, B, D). Fourth, the lower clay content in the anthropic ADE horizons at Caldeirão (erroneously described by Silva et al. as “sandy clay loam”5) is not evidence of fluvial deposition but a partial outcome of argilluviation9. Fifth, other well-studied ADE sites nearby contradict Silva et al.’s inference5: at the Hatahara ADE site, located 4 km from Caldeirão on the same Miocene bluff, the similarity in quartz sand grain morphology between the ADE A and B horizons excludes the inference of fluvial inputs into the A horizon13. Further afield, a large number of ADE sites are found along blackwater (non alluvial) rivers, associated with small headwater streams and springs, or found at elevations exceeding 90 m above the maximum flood level14–16, demonstrating that alluvial deposition is irrelevant to the formation of many ADE expanses. Indeed, if ADE were the result of alluvial processes, their spatial distribution along rivers would be continuous rather than patchy. 2 Archaeological considerations Research conducted at numerous archaeological sites in the Central Amazon17 has shown that the largest ADE expanses record multicomponent occupations that date to the period 1200–800 BP and are often underlain by remains of older (<2500 BP) ceramic occupations2–4,6. This also applies in the case of the Caldeirão site, where coring and excavations clearly show that the ADE is a pottery-rich archaeological deposit characterised by a predominantly human-made assemblage of mounds and pits (Fig. 2A–E). Silva et al.’s sampling transects and elemental/isotopic measurements neither take into consideration nor detect this demonstrable anthropic conditioning of pre-Columbian origin (see Inset II in Fig. 2E)5. Furthermore, Silva et al. misunderstand stratigraphic associations when suggesting that >7.6 ky 14C BP charcoal collected from −90 cm in their Ultisol transect provides an accurate age marker for the beginning of ADE formation5. Middle Holocene charcoal fragments are commonly found stratified in Amazonian soil profiles18, including the B horizons of ADE profiles14. However, the relevant age to understand ADE formation (and whether it is consistent with human occupation) is that of the silt-sized charcoal making up the dark horizon of an ADE. At the nearby ADE site of Hatahara, the age of this charcoal pool is consistent with a late first millennium AD Paredão phase settlement, albeit with older occupations starting around 500 BC19,20. For Caldeirão, similar ages have been reported21. NATURE COMMUNICATIONS | (2022)13:3444 | https://doi.org/10.1038/s41467-022-31064-2 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-022-31064-2 Demographic considerations Silva et al. argue that a late Holocene onset for incipient agriculture in the Central Amazon region would preclude populations large enough to produce the levels of elemental enrichment recorded at Caldeirão5. This argument presupposes that indigenous land use regimes relying on incipient agriculture, aquatic wildlife, and hunting could not have created areas of persistent high fertility. This MATTERS ARISING assumption does not account for decades of research on the subject. For instance, ethnoarchaeological research with the Kuikuro community, who are fisher-cultivators that live in the Upper Xingu region, has demonstrated that the greatest enrichment in P, Ca, and Sr, as well as high organic carbon and nearly neutral pH, occurs in mounded refuse middens. Once enriched soil horizons form in the middens, typically within a few years, they are often used for NATURE COMMUNICATIONS | (2022)13:3444 | https://doi.org/10.1038/s41467-022-31064-2 | www.nature.com/naturecommunications 3 MATTERS ARISING NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-022-31064-2 Fig. 2 Archaeological fieldwork—excavations and mapping—carried out at the Caldeirão site in 2011. A, B, C, and D Vertical profiles exposed by multiple archaeological excavations at the Caldeirão ADE. E Google Earth image of the Caldeirão ADE (see location of profiles A–D within insets I and II). 2a and 2b are ~25 m apart and show the stratigraphy of archaeological deposits in mound (2a) and flat (2b) areas. 2c and 2d are ~12 m apart and show the stratigraphy of archaeological deposits at an Embrapa reference profile (C) and nearby archaeological excavation (D). Note clearly defined archaeological matrix features infilled with ADE sediment (C), and infilled pit feature with well-preserved ceramic vessels, suggesting intentional deposition by ancient indigenous Amazonians (D). E Yellow shaded area shows the spatial distribution of mounds and archaeological pottery ascertained through archaeological survey and excavation. Insets I, II, II show details of the topography and/or archaeological excavations, as well as sampling location for profiles depicted in (A–D). Inset II: Note the close proximity between identified mounded areas (black arrows), archaeological excavations, and the area of the ADE sampled by Silva5 (blue rectangle). Inset III: Survey has also identified mounded areas (black arrows) near the area Silva et al.5 sampled for Ultisols (red rectangle). cultivating crops such as maize, sweet potato, and manioc22. Soil enrichment and ADE formation, therefore, are consistently associated with domestic activities in indigenous villages and, contrary to Silva et al.’s claim5, it is this elemental enrichment accumulating in settlements that is used for cultivation (and not the other way around). More broadly, measurements of elemental enrichment with P and Ca constitute a poor demographic proxy and, on their own, do not reveal agricultural activity: virtually any long human occupation can result in soil enrichment23. ADE sites, like Caldeirão, are very rich in nutrients because they concentrate human debris and waste associated with resources gathered or produced in large areas. It is the concentration of resources in settlements that produce ADEs over hundreds or thousands of years. Put another way, a thousand people could extract resources produced from a 50 hectares’ catchment but concentrate debris and waste in a village of 0.1 hectares. Silva et al.’s5 reference to improbably large agricultural populations, which implicitly suggests that ADEs were initially established for agricultural purposes, does not constitute evidence of fluvial deposition and disregards the association between ADE and middens that is supported by current research. ratio in ADEs is the result of differential preservation coupled with the specific tropical soil dynamics of Ca, which is easily leached, and P, which binds with soil Fe and Al oxides28. By way of conclusion: the geogenic model for ADE formation, which famously argued that ADEs are dark soils of natural fertility resulting from the deposition of alluvial horizons10, was laid to rest over 40 years ago29. Silva et al.’s hypothesis5 reiterates this geogenic position but, as we have shown here, it does not stand up to scrutiny. Data availability All relevant data are provided with the paper. Received: 21 January 2021; Accepted: 27 May 2022; References 1. Elemental enrichment and isotopic ratios of ADE vs. Ultisols (Acrisols) Most of the co-authors of Silva et al.5 have elsewhere argued that the elemental composition of Caldeirão site “…can be used to unveil ADE sites and differentiate them from Amazonian soils without anthropic influence”24. We agree with their earlier assessment: enrichment of the ADE compared to the Ultisols is consistent with inputs associated with human settlement. Among the latter are those related to burning, including K, Rb, Ba, Ca, Sr, P (from ash and charcoal); P, Ca, Sr, K, Zn, Cu (human waste); and Ca, P, Sr, Zn (bone debris) (Fig. 1B, C)25. Most of these, along with pyrogenic C, have been reported in ADEs8. The most logical explanation for such an assemblage is anthropic inputs associated with settlement activity. Indeed, research at the Hatahara site shows that the dark ADE sediments are bulked up by sand and silt-sized particulate material resulting from anthropic activity (fragmented charcoal and bone, pottery fragments, sponge spicules, etc.)13. Bioturbation can then mix these added materials in soil over time throughout the profile. How, then, can a fluvial input be surmised? The core of Silva et al.’s argument is that differences in Sr and Nd isotope ratios between ADE and Ultisols are best explained by fluvial inputs5. However, both Sr and Nd are found in plants26 and terrestrial and aquatic vertebrates27, as well as in mineral matter and Silva et al. admit that their methods cannot discriminate these sources5. As there are no independent indications of sediment input in ADE’s bulk chemical composition, but ample evidence for non-mineral anthropogenic inputs, it is most likely that isotopic signature in the studied ADE resulted from the deposition of food debris. Silva et al. regard the difference in elemental stoichiometries of freshwater fish (Ca:P ~2.13) and human faeces (Ca:P ~2) compared with ADEs as further evidence of ADE being of fluvial origin5. However, while the Ca:P ratio is highly variable in Caldeirão ADE (Fig. 1C), the modern Ca:P 4 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Clement, C. R. et al. The domestication of Amazonia before European conquest. Proc. R. Soc. Lond. B Biol. Sci. 282 https://doi.org/10.1098/rspb.2015.0813 (2015). Neves, E. G., Guapindaia, V. L. C., Lima, P. H., Costa, B. L. S. & Gomes, J. in Amazonía. Memorias de las Conferencias Magistrales del 3er Encuentro Internacional de Arqueología Amazónica (ed. Rostain, S.) 137–157 (Ministerio Coordinador de Conocimiento y Talento Humano/IKIAM, 2014). Moraes, Cd. P. & Neves, E. G. O ano 1000: Adensamento populacional, interação e conflito na Amazônia Central. Amazonica 4, 122–148 (2012). Arroyo-Kalin, M. & Riris, P. Did pre-Columbian populations of the Amazonian biome reach carrying capacity during the Late Holocene? Philos. Trans. R. Soc. Lond. B Biol. Sci. 376, 20190715 (2021). Silva, L. C. R. et al. A new hypothesis for the origin of Amazonian Dark Earths. Nat. Commun. 12, 127 (2021). Schmidt, M. J. et al. Dark earths and the human built landscape in Amazonia: a widespread pattern of anthrosol formation. J. Archaeological Sci. 42, 152–165 (2014). Arroyo-Kalin, M. In Archaeological Soil and Sediment Micromorphology (eds Georges Stoops & Cristiano Nicosia) 345–357 (Wiley and sons, 2017). Glaser, B. & Birk, J. J. State of the scientific knowledge on properties and genesis of Anthropogenic Dark Earths in Central Amazonia (terra preta de Índio). Geochim. Cosmochim. Acta 82, 39–51 (2012). Macedo, R. S., Teixeira, W. G., Corrêa, M. M., Martins, G. C. & Vidal-Torrado, P. Pedogenetic processes in anthrosols with pretic horizon (Amazonian Dark Earth) in Central Amazon, Brazil. PLoS ONE 12, e0178038 (2017). Falesi, I. C. In Man in the Amazon (ed. Wagley, C.) 201–229 (The University Presses of Florida, 1974). Pupim, F. N. et al. Chronology of Terra Firme formation in Amazonian lowlands reveals a dynamic Quaternary landscape. Quat. Sci. Rev. 210, 154–163 (2019). Guyot, J. L. et al. Clay mineral composition of river sediments in the Amazon Basin. Catena 71, 340–356 (2007). Arroyo-Kalin, M., Neves, E. & Woods, W. I. In Amazonian Dark Earths: Wim Sombroek’s Vision (eds Woods, W. I. et al.) 99–125 (Springer Netherlands, 2009). Heckenberger, M. J., Petersen, J. B. & Neves, E. G. Village size and permanence in Amazonia: two archaeological examples from Brazil. Lat. Am. Antiquity 10, 353–376 (1999). Guapindaia, V. & Aires Da Fonseca, J. Metodologia de delimitação no sítio arqueológico Cipoal do Araticum na região do rio Trombetas, Pará, Brasil. Bol. Mus. Para. Emílio Goeldi Ciênc. Humanas 8, 657–673 (2013). NATURE COMMUNICATIONS | (2022)13:3444 | https://doi.org/10.1038/s41467-022-31064-2 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-022-31064-2 16. Eden, M. J., Bray, W., Herrera, L. & McEwan, C. Terra Preta Soils and Their Archaeological Context in the Caqueta Basin of Southeast Colombia. Am. Antiquity 49, 125–140 (1984). 17. Neves, E. G. In Ethnicity in ancient Amazonian: reconstructing past identities from Archaeology, Linguistic and Etnohistory 1–27 (University Press of Colorado, 2011). 18. Pessenda, L. C. R. et al. 14C Dating and stable carbon isotopes of soil organic matter in forest–Savanna Boundary Areas in the Southern Brazilian Amazon Region. Radiocarbon 40, 1013–1022 (1997). 19. Arroyo-Kalin, M. Slash-burn-and-churn: landscape history and crop cultivation in pre-Columbian Amazonia. Quat. Int. 249, 4–18 (2012). 20. Neves, E. G., Petersen, J. B., Bartone, R. N. & Heckenberger, M. J. In Amazonian Dark Earths: Explorations in Space and Time (eds Glaser, B. & Woods, W. I.) 125–134 (Springer Berlin Heidelberg, 2004). 21. Schellekens, J. et al. Molecular composition of several soil organic matter fractions from anthropogenic black soils (Terra Preta de Índio) in Amazonia —a pyrolysis-GC/MS study. Geoderma 288, 154–165 (2017). 22. Schmidt, M. Amazonian Dark Earths: pathways to sustainable development in tropical rainforests? Bol. do Mus. Para. Emílio Goeldi. Ciências Humanas 8, 11–38 (2013). 23. Canti, M. & Huisman, D. J. Scientific advances in geoarchaeology during the last twenty years. J. Archaeol. Sci. 56, 96–108 (2015). 24. Barbosa, J. Z. et al. Elemental signatures of an Amazonian Dark Earth as result of its formation process. Geoderma 361, 114085 (2020). 25. Oonk, S., Slomp, C. P. & Huisman, D. J. Geochemistry as an aid in archaeological prospection and site interpretation: current issues and research directions. Archaeol. Prospection 16, 35–51 (2009). 26. Tyler, G. Rare earth elements in soil and plant systems—a review. Plant Soil 267, 191–206 (2004). 27. Tütken, T., Vennemann, T. W. & Pfretzschner, H.-U. Nd and Sr isotope compositions in modern and fossil bones – Proxies for vertebrate provenance and taphonomy. Geochim. Cosmochim. Acta 75, 5951–5970 (2011). 28. Sato, S., Neves, E. G., Solomon, D., Liang, B. Q. & Lehmann, J. Biogenic calcium phosphate transformation in soils over millennial time scales. J. Soils Sediment. 9, 194–205 (2009). 29. Smith, N. J. H. Anthrosols and human carrying capacity in Amazonia. Ann. Assoc. Am. Geographers 70, 553–566 (1980). Author contributions U.L., M.A.K., M.J.S., H.H., H.P.L., C.P.M., E.G.N., W.T., and C.R.C. co-wrote the paper with inputs from J.A.F., F.O.A., C.B.V.A., C.B.R., G.G.B., M.S.C., M.L.C., L.C., L.H.C.A., W.M.D., C.F., C.F.C., A.F., B.F., B.G., M.H., S.H., V.H., K.A.J., A.B.J., T.K., E.K.T., MATTERS ARISING T.W.K., J.L., M.M., S.Y.M., L.M.C., F.E.M., D.M., B.M., G.M.R., C.A.P.B., M.P.M., G.P.C., F.P., F.N.P., M.F.R., A.R.P.D., P.R., B.R., L.R., S.R., R.S.M., M.P.S., T.S., F.S.B., R.V., P.V.T., X.S.V., J.W., and S.L.W. H.H. prepared Fig. 1. H.P.L. and M.J.S. prepared Fig. 2A–D. M.A.K., M.J.S., and W.T. prepared Fig. 2E with input from H.P.L., E.G.N., C.M., and U.L. H.P.L., B.M., E.G.N., M.J.S., C.F.C., G.P.C., F.S.B., M.S.C. and R.V. carried out archaeological fieldwork at Caldeirão. Competing interests The authors declare no competing interests. Additional information Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41467-022-31064-2. Correspondence and requests for materials should be addressed to Umberto Lombardo or Manuel Arroyo-Kalin. Peer review information Nature Communications thanks William Balée, Talitha Santini and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Reprints and permission information is available at http://www.nature.com/reprints Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/ licenses/by/4.0/. © The Author(s) 2022 1 Institut de Ciència i Tecnologia Ambientals, Universitat Autònoma de Barcelona (ICTA-UAB), Bellaterra, Barcelona, Spain. 2Geographical Institute, University of Bern, Bern, Switzerland. 3Institute of Archaeology, University College London, London, UK. 4Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA, USA. 5Groningen Institute of Archaeology, University of Groningen, Groningen, Netherlands. 6Museu Paraense Emílio Goeldi, Belém, Brazil. 7Instituto de Ciências da Sociedade, Universidade Federal do Oeste do Pará, Santarém, Brazil. 8Museum of Archaeology and Ethnology, University of São Paulo, São Paulo, Brazil. 9Instituto Nacional de Pesquisas da Amazônia, Manaus, Brazil. 10ArqueoMaquina, Belém, Brazil. 11Departamento de Arqueologia, Universidade do Estado do Rio de Janeiro, Rio de Janeiro, Brazil. 12Wageningen University & Research, Wageningen, Netherlands. 13School of Archaeology, University of Oxford, Oxford, UK. 14 Embrapa Forestry, Colombo, Brazil. 15Geosciences Institute, Federal University of Pará, Belem, Brazil. 16Centro de Ecologia Funcional, Universidade de Coimbra, Coimbra, Portugal. 17Soils Department, Federal Rural University of Rio de Janeiro, Seropédica, Brazil. 18Department of Geography, University of Wisconsin-Madison, Gualala, CA, USA. 19Museu Nacional, Universidade Federal do Rio de Janeiro, São Cristóvão, Brazil. 20 Princeton Institute for International and Regional Studies, Princeton University, Princeton, NJ, USA. 21Naturalis Biodiversity Center, Leiden, Netherlands. 22Embrapa Solos, Rio de Janeiro, Brazil. 23Department of Soil Biogeochemistry, Martin-Luther-Universität Halle-Wittenberg, Halle, Germany. 24Department of Anthropology, University of Florida, Gainesville, FL, USA. 25School of Public Affairs, UCLA, Los Angeles, CA, USA. 26Graduate Institute for International Development Research, Geneva, Switzerland. 27Instituto de Desenvolvimento Sustentável Mamirauá, Tefé, Brazil. 28School of Integrative Plant Science, Department of Global Development, Cornell University, Ithaca, NY, USA. 29Culture and SocioEcological Dynamics Research Group, Department of Humanities, Universitat Pompeu Fabra, Barcelona, Spain. 30Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain. 31Department of Ecosystem and Landscape Dynamics, University of Amsterdam, NATURE COMMUNICATIONS | (2022)13:3444 | https://doi.org/10.1038/s41467-022-31064-2 | www.nature.com/naturecommunications 5 MATTERS ARISING NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-022-31064-2 Amsterdam, Netherlands. 32Department of Archaeology, Max Planck Institute for the Science of Human History, Jena, Germany. 33Faculty of Archaeology, Leiden University, Leiden, Netherlands. 34Department of Geography and Environmental Science, University of Reading, Reading, UK. 35 CEFE, Univ Montpellier, CNRS, EPHE, IRD, Univ Paul-Valéry Montpellier, Montpellier, France. 36Amazon Hopes Collective, Belém, Brazil. 37 Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogotá, Colombia. 38Institute of National Historic and Artistic Heritage, Belém, Brazil. 39Departamento de Ciências Ambientais, Universidade Federal de São Paulo, Diadema, Brazil. 40Institute for Modelling SocioEnvironmental Transitions, Bournemouth University, Poole, UK. 41Climate and Agriculture Group, Agroscope, Zurich, Switzerland. 42French National Centre for Scientific Research, Paris, France. 43Instituto Nacional do Semiárido (INSA), Campina Grande, Brazil. 44Center of Competence for Soils, BFH-HAFL, Zollikofen, Switzerland. 45Museu da Amazônia, Manaus, Brazil. 46Soil Science Department, University of São Paulo, Piracicaba, Brazil. ✉email: Umberto.Lombardo@uab.cat; m.arroyo-kalin@ucl.ac.uk 6 NATURE COMMUNICATIONS | (2022)13:3444 | https://doi.org/10.1038/s41467-022-31064-2 | www.nature.com/naturecommunications