Résumé
Following the MISTRALS Framework, the PALEOMEX program primarily aims to reconstitute past environments. In this study, we present a reconstitution of the vegetation of the Languedoc-Roussillon territory on the Mediterranean shores of France during the pre-neolithic times using climate and soil habitats of various Plant Functional Types (PFTs). We developed an agent-based model of the vegetation and biotope and validated the result with LPJ-Guess which utilizes the European Pollen Database. To reconstitute the potential vegetation biocenosis of the time, we first classified the 94 tree and bush species according to archeological records into 10 PFTs. Here, we based the classification on the three most cited variables affecting the species’ behavior: water availability, mean annual temperature, and soil pH. Next, we determined the spatial characteristic of the territory by simulating the Roussillon territory with its biotope. The following are considered in the simulation: climate (rainfall and temperature from WorldClim), elevation and pedology (pH and fertility), and water dynamics (run-off, river fluxes, and water table balance). Lastly, we considered the probability of extreme events (or, accidents) based on local temperature and water availability (where temperature variability directly relates to the probability of accidents). In this study, we considered frosts, tempests, floods, and fires - all directly affecting the state and renewal of vegetation cover and PFTs. The model is run on a 90m*90m elevation grid: on each pixel, water availability and PFT evolution are simulated at seasonal temporal resolution (using climate and soil data). The evolution of the landscape is simulated over a period of 500 years. Initialization goes through the succession of average seasons, including extreme events. As a result, we present a dynamic reconstitution of vegetation and biotopes that can be tested against archeological data and used for the next step (that is, the implementation of Neolithic farmers and herders on this reconstituted Roussillon territory). Lastly, we plan to formalize the various socio-anthropological constraints of the residing Neolithic societies, considering the complexity, robustness, and expansion of family organizations. The main contribution of this work is the vegetation cover of the Roussillon territory during the pre-neolithic period based on the PFT distribution.