Modeling initial Neolithic dispersal. The first agricultural groups in West Mediterranean

• Published in: Ecological modelling Vol. 307; pp. 22 - 31
• Main Authors: Bernabeu Aubán, Joan, Michael Barton, C., Pardo Gordó, Salvador, Bergin, Sean M.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01-07-2015
• Subjects: Agent based models; Agronomy; Complex Adaptive Systems; Dynamics; Ecology; Economics; Empirical analysis; Farming; Farms; Neolithic spread; Socio-ecological modeling; Stone Age; West Mediterranean Neolithic; ANE, Europe, Iberian Peninsula; MED, Western Mediterranean; MED; ANE, Europe; Neolithic spread; Complex Adaptive Systems; Socio-ecological modeling; Agent based models; West Mediterranean Neolithic
• ISSN: 0304-3800; 1872-7026
• DOI: 10.1016/j.ecolmodel.2015.03.015

•We explore the Neolithic spread as an emergent phenomenon resulting from local-scale processes.•We use an agent-based model and four c14 data sets to evaluate model results.•We test the effects of Ecology, Demography, Movement and Dispersal Routes.•Results emphasize the use of c14 data set based only on dating of domestic taxa.•Best results include N & S arrivals on unsettled regions with highest land quality.
In previous research, the SE-NW time-trend in the age of the earliest Neolithic sites across Europe has been treated as a signal of a global-scale process that brought farming/herding economies to the continent. Residual variation from this global time-trend is generally treated as ‘noise’. A Complex Adaptive Systems perspective views this empirical record differently. The apparent time-trend is treated as an emergent consequence of the interactions of individuals and groups of different scale.
Here, we examine the dynamics of agricultural dispersals, using the rich body evidence available from the Iberian Peninsula as a case study. We integrate two complementary approaches: (1) creating a high resolution Agent Based Modeling environment to simulate different processes that may have driven the spread of farming; (2) collecting and synthesizing empirical archeological data for the earliest Neolithic settlements that we use to evaluate our models results.
Our results suggest that, (a) the source of radiocarbon data used to evaluate alternative hypotheses play an important role in the results; and (b) the model scenario that produces de best fit with archeological data implies a dispersal via northwestern and southern routes; a preference for leap-frog movement; an influence of ecological conditions (selecting most favorable agricultural land) and demographic factors (avoiding settled regions).
This work represents a first attempt at high-resolution bottom-up modeling of this important dynamic in human prehistory. While we recognize that other social and environmental drivers could have also affected the dispersal of agropastoral systems, those considered here include many that have been widely considered important in prior research and so warrant inclusion.