Modelling the ancient maize agriculture potential of landforms in Tikal National Park, Guatemala

The ancient polity of Tikal has been extensively studied by archaeologists and soil scientists, but more information is needed to determine the specific subsistence and ancient farming techniques that sustained its inhabitants for more than eight centuries. Recent settlement, soil resource, and vegetation surveys were completed during a re-evaluation of the earthworks of Tikal. The objective of this research was to combine non-parametric multiplicative regression, stable carbon isotopes, soil properties, and remote-sensing data for cost-effective, predictive modelling of ancient maize (Zea mays L.) agricultural areas. The study area is positioned within a 300 km(2) region surrounding the Tikal site centre (17 degrees 13 19.0 N, 89 degrees 27 25.2 W). In the study area, ancient maize agriculture was determined using carbon (C) isotopic signatures of C-4 vegetation that were incorporated into the soil humin fraction. Probability models predicting C isotopic enrichment were used to outline areas of potential long-term maize agriculture. A binomial model predicted areas mainly along the bajo margins, indicating that these areas may have been used for sustained ancient maize agriculture. Upland areas with shallow Rendolls soil were found to lack strong C-4 isotopic signatures. They showed a long-term C-3 vegetation signature that could have supported native forest, silvicultural activities, or C-3 cropping practices (i.e. forest gardens). The mosaic of C-3 and C-4 signatures within the study site soils suggests that the Tikal area supported a variety of food production systems.