Friday 17 January 2014

“Predicting pre-Columbian anthropogenic soils in Amazonia”

ResearchBlogging.org “Predicting pre-Columbian anthropogenic soils in Amazonia” is the title of a recent paper by McMichael et al. published in Proceeding of the Royal Society B. It is not open access but you can read the abstract here. In this paper McMichael et al. present a predictive model for the presence of Terra Preta in Amazonia.  The model predicts the likelihood of finding Terra Preta sites in any given spot within Amazonia.  In general, I liked the idea behind the paper. These models give us an objective basis for further research and discussions. It is thanks to this kind of work that we can go beyond subjective views about the extent of human impact in pre-Columbian Amazonia and start to formulate hypothesis that, through survey and measurement, can be later tested. The first important result of this paper is that, given the data available, we can now estimate that terra preta is likely to be found only in a 3.2% of the forested areas of Amazonia. This is far less than other previous estimates (Erickson, 2008).

However, I find the discussion of McMichael et al. a bit disappointing with regards to two points. The first is the meaning they give to the presence of terra preta; the second is the reasons they give to explain the absence terra preta outside of Brazil.
What is terra preta? Is it the result of permanent settlement where people cooked and dumped food remains for centuries, eventually causing the enrichment of the soil in with charcoal, phosphorous, organic matter and the rest of elements used to define terra preta? Or is it the result of soil management techniques aimed at improving fertility and agricultural potential? McMichael et al. seem to imply that terra preta is the latter: the result of soil fertility enhancement. They say: “The lack of terra pretas in western Amazonia may be because the Andean-derived soils of western Amazonia did not require nutrient enrichment… [the bold is mine]”. However, this kind of interpretation of terra preta being the result of Pre-Columbian agricultural intensification has been challenged by many authors. Neves & Petersen (2006) discovered that at the Hatahara occupation site (close to Manaus) pre-Columbians actually used terra preta to build burial mounds, which is a strange use for an agricultural soil that took centuries to form. Of course, we cannot exclude that pre-Columbians took advantage of the fertile terra preta for their gardens associated to their homes; in the same way that they could have taken advantage of the fertility of pre-existing middens (see Arroyo-Kalin, 2012 for a discussion on this).But this does not mean that people intentionally created terra preta for agriculture!
Recently, Glaser & Birck (2012) concluded their review about the state of the scientific knowledge about the properties and genesis of Anthropogenic Dark Earths in Central Amazonia saying: “there is no scientific evidence indicating that forgotten agricultural techniques for large scale soil fertility improvement are responsible for terra preta genesis”. 

This leads us to my second concern: what does it mean when no terra preta is found? Here, McMichael et al. suggest that the lack of terra preta indicates that people decided to produce food in some other way, due to cultural and/or environmental reasons. They say: “[In the Llanos de Moxos] instead of terra preta formation, large societies sustained themselves by using techniques such as fish weirs and raised-field agriculture”. But, is it cultivating little gardens that large societies sustained themselves? I think the answer is no. In fact, pre-Columbians living in terra preta sites performed agriculture in the surrounding area, eventually forming terra mulata sites. Terra mulata sites are far larger than terra preta ones. Terra mulata sites do not contain pottery and are far less fertile than terra preta ones, but still, they are richer in organic matter than the normal Amazonian oxisols (more on this here). It is terra mulata that formed because of ancient agricultural use, not terra preta.
The main problem we face when tackling the question of terra preta is its definition. The definition that is generally given to terra preta coincides with the description of the geochemistry of a midden (From Wikipedia: an old dump for domestic waste which may consist of animal bone, human excrement, botanical material, vermin, shells, sherds, lithics, and other artifacts and ecofacts associated with past human occupation) and there are middens everywhere in the world! As a consequence of this, terra preta sites are now appearing everywhere... We should also keep in mind that the whole terra preta concept is rooted in the context of the big surprise that the first researchers had when they found black organic sediments in the middle of Amazonian heavily weathered soils. In fact, terra preta is often defined (and mapped in the field) in relation to the surrounding soil (Fig. 1). In my view, there is not much that actually differentiates terra preta from other occupation horizons elsewhere. I have seen several places in the Bolivian Amazon that, because of the colour of the soil, concentrations of P, Ca, charcoal etc., would fit quite well into the definition of terra preta (see for example this). It is just that they are not called terra preta, yet :- ). It could be that the absence of terra preta sites outside Brazil is merely the result of researchers giving these kinds of soils/deposits different names in different regions, such as “middens” or “occupation horizons”.

Figure 1: oxisol left, terra preta right (from Wikipedia)
I think that the, otherwise excellent, paper by McMichael at al. could have benefited from incorporating into their model a database of terra mulata sites, instead of terra preta sites. Or, even better, if they had used a database of pre-columbian occupations, including the archaeological sites known outside of Brazil. This would have provided a more reliable tool for modelling pre-Columbian agriculture (if a terra mulata database had been used) or settlements patterns (if a database of archaeological sites had been used) within the Amazon basin; and for modelling pre-Columbian disturbance of the natural environment.

References
 McMichael CH, Palace MW, Bush MB, Braswell B, Hagen S, Neves EG, Silman MR, Tamanaha EK, & Czarnecki C (2014). Predicting pre-Columbian anthropogenic soils in Amazonia. Proceedings. Biological sciences / The Royal Society, 281 (1777) PMID: 24403329

Erickson, C.L. (2008). Amazonia: the historical ecology of a domesticated landscape In: H. Silverman, W.H. Isbell (Eds.), Handbook of South American archaeology. Springer, Berlin, pp. 157-183 DOI: 10.1007/978-0-387-74907-5_11

Neves, E.G., & Petersen, J.B. (2006). Political economy and pre-Columbian landscape transformations in Central Amazonia In: W. Balée & C.L. Ericksonl (Eds.),Time and complexity in historical ecology. Columbia University Press, New York, pp. 279-309

Arroyo-Kalin, M. (2012). Slash-burn-and-churn: Landscape history and crop cultivation in pre-Columbian Amazonia Quaternary International, 249, 4-18 DOI: 10.1016/j.quaint.2011.08.004

Glaser, B., & Birk, J.J. (2012). State of the scientific knowledge on properties and genesis of Anthropogenic Dark Earths in Central Amazonia (terra preta de Índio) Geochimica et Cosmochimica Acta, 82, 39-51 DOI: 10.1016/j.gca.2010.11.029