Nutrient status in prehistoric agricultural soils – indications from geochemical analyses of archaeological topsoil relicts

Abstract

At the beginning of agriculture in Central Europe during the Neolithic (at about 5,500 BC) first farmers started to utilise the fertile loess areas. However, archaeologists have shown that the location shifted between settlement areas and, therefore, also between agricultural fields. Besides cultural and other environmental variables, it has been assumed that decreasing soil fertility was one key factor leading to human mobility. To test this hypothesis, the soil nutrient status and organic matter origin in Neolithic and Bronze Age topsoils were investigated. For this purpose, buried prehistoric topsoil relicts from loess areas in Central Germany (now comprising Chernozems and Phaeozems) and in Western Germany (now comprising Luvisols) were sampled from archaeological excavations. These relicts were preserved as pit fillings in slot pits and pit alignments outside prehistoric settlements. Such features are part of the prehistoric agrarian landscape, and therefore their dark humic soil filling consists to a large extent of ancient topsoil from arable land. Furthermore, in the West German study region four colluvial deposits ranging from the Bronze Age until recent times were analysed for diachronic comparisons. To reconstruct the soil fertility status of prehistoric arable soils, geochemical analyses were conducted comprising (i) a general soil characterisation (including soil colour, organic carbon and total nitrogen contents, pH, and grain size distribution), (ii) a Hedley fractionation of soil phosphorus (P), (iii) total contents of selected micronutrients, (iv) 15N isotopy, and (v) different biomarkers. The latter consisted of amino acids to elucidate the crop origin of proteins, of steroids to identify faecal residues, and of benzene-carboxylic acids to trace charred plant remains.
The prehistoric topsoils in the pit fillings contained more organic matter and P than the adjacent subsoils, underlining their former topsoil character. The extractable P content in the pit fillings even amounted to 86±4% of the extractable P contents of the recent, fertilised topsoils. The dominating P form was inorganic P (>80%); yet, plant available P contents (i.e. sum of resin-Pi, NaHCO3-Pi and NaOH-Pi) in the prehistoric and recent topsoils were similar (23% and 20% of extractable P, respectively), suggesting that even plant availability of P was maintained in the prehistoric topsoil relicts. Hence, the hypothesis that there was a nutrient deficiency in prehistoric agriculture has to be refuted, at least as P is concerned as one important macronutrient. Also micronutrient analyses of I, Cu, Mn, Mo, Se and Zn did not point to any nutrient deficiencies in the prehistoric topsoils.
In the colluvial deposits, the reconstruction of the soil P status of different prehistoric time periods was challenging due to varying pre- and post-depositional factors and transformations. Nevertheless, Bronze/Iron Age colluvial layers revealed large extractable P contents as well as a high P availability and confirmed, thus, a sufficient nutrient supply as indicated by the pit investigation.
The organic matter of the prehistoric topsoils showed pronounced enrichment of the heavy N isotope, leading to δ15N values close to 7‰ as commonly found in manured topsoils. The analyses of bile acids finally confirmed that the prehistoric topsoils must have received an input of livestock manure. There was no unambiguous evidence of prehistoric legume cultivation or potential N inputs via green manuring using amino acid analyses. However, black carbon contents of up to 38% of soil organic C and the dark soil colour in the pit fillings provided clear evidence of prehistoric burning events and charcoal input. The black carbon quality indicated rather hot burning temperatures and therefore a human-induced charcoal input into the soils is likely.
In conclusion, the combined analyses of different geochemical proxies made it possible to reveal a good soil P and micronutrient status of prehistoric arable soils. On the one hand, prehistoric population density was probably too low to overuse the fertile loess soils. On the other hand, there are several hints for enhanced inputs of organic material like charcoal and manure from livestock faeces. Therefore, soil nutrient depletion of Neolithic and Bronze Age agricultural soils was likely not the reason for the abandonment of arable fields and human mobility in the study area.

Resources

Bibliography

Lauer, F. (2014): Nutrient status in prehistoric agricultural soils – indications from geochemical analyses of archaeological topsoil relicts. University of Bonn