Traditional GIS-based statistical models are intended to extrapolate patterns of settlements and their interactions with the environment. They contribute significantly to our knowledge of past human–land relationships. Yet, these models are often criticized for their empiricism, lopsided specific factors, and for overlooking the synergy between variables. Though largely untested, machine learning and artificial intelligence methods have the potential to overcome these shortcomings comprehensively and objectively. The northeastern Tibetan Plateau (NETP) is characterized by diverse environments and significant changes to the social system from the Neolithic to Bronze Age. In this study, this area serves as a representative case for assessing the complex relationships between settlement locations and geographic environments, taking full advantages of these new models. We have explored a novel modeling case by employing GIS and random forests to consider multiple factors, including terrain, vegetation, soil, climate, hydrology, and land suitability, to construct classification models identifying environmental variation across different cultural periods. The model exhibited strong performance and a high archaeological prediction value. Potential living maps were generated for each cultural stage, revealing distinct environmental selection strategies from the Neolithic to Bronze Age. The key environmental parameters of elevation, climate, soil erosion, and cultivated land suitability were calculated with high weights, influencing human environmental decisions synergistically. Furthermore, we conducted a quantitative analysis of temporal dynamics in climate and subsistence to understand driving mechanisms behind environmental strategies. These findings suggest that past human environmental strategies were based on the comprehensive consideration of various factors, coupled with their social economic scenario. Such subsistence-oriented activities supported human beings in overcoming elevation limitation, and thus allowed them to inhabit wider pastoral areas. This study showcases the potential of machine learning in predicting archaeological probabilities and in interpreting the environmental influence on settlement patterns.

The role of human activity in shaping the terrestrial environment has been a core scientific issue of interest across various disciplines. However, it remains unclear whether there are significant differences in the patterns of the anthropogenic impact on the terrestrial environment in terms of spatial and temporal dimensions, and we are yet to identify the underlying factors that have driven it. Here, we present an analysis of sporopollen and geochemical proxies from a section of the Anjiangbei site (AJB) on the Yunnan Plateau, spanning the Ming–Qing period, in order to explore the spatio-temporal variation in the anthropogenic impact on the terrestrial environment in the Lake Dian basin. Integrating the reported multidisciplinary evidence, we aim to reveal the influencing factors of anthropogenic impact. Our results show that there were remarkable differences in anthropogenic impact on the terrestrial environment in the Lake Dian basin between the Late Bronze Age and the Ming–Qing period. Changes in crop vegetation and the forest were all affected by human activity in the Lake Dian basin during the two periods, and were more evident during the Ming–Qing period. The heavy metal pollution in the soil was obvious during the Ming–Qing period. The increase in the intensity of human activity, especially the rise in population, could be attributed to changes in the hydrological environment in the Lake Dian basin during the Late Bronze Age and to geopolitical change during the Ming–Qing period. This study reveals the different patterns in human impact on the terrestrial environment in the Lake Dian basin during the Late Bronze Age and the Ming–Qing period, providing new evidence to enable a deeper understanding of past human–environment interactions on the Yunnan Plateau.

Traditional GIS-based statistical models are intended to extrapolate patterns of settlements and their interactions with the environment. They contribute significantly to our knowledge of past human–land relationships. Yet, these models are often criticized for their empiricism, lopsided specific factors, and for overlooking the synergy between variables. Though largely untested, machine learning and artificial intelligence methods have the potential to overcome these short-comings comprehensively and objectively. The northeastern Tibetan Plateau (NETP) is character-ized by diverse environments and significant changes to the social system from the Neolithic to Bronze Age. In this study, this area serves as a representative case for assessing the complex re-lationships between settlement locations and geographic environments, taking full advantages of these new models. We have explored a novel modeling case by employing GIS and random forests to consider multiple factors, including terrain, vegetation, soil, climate, hydrology, and land suitability, to construct classification models identifying environmental variation across different cultural periods. The model exhibited strong performance and a high archaeological prediction value. Potential living maps were generated for each cultural stage, revealing distinct environ-mental selection strategies from the Neolithic to Bronze Age. The key environmental parameters of elevation, climate, soil erosion, and cultivated land suitability were calculated with high weights, influencing human environmental decisions synergistically. Furthermore, we conducted a quan-titative analysis of temporal dynamics in climate and subsistence to understand driving mecha-nisms behind environmental strategies. These findings suggest that past human environmental strategies were based on the comprehensive consideration of various factors, coupled with their social economic scenario. Such subsistence-oriented activities supported human beings in over-coming elevation limitation, and thus allowed them to inhabit wider pastoral areas. This study showcases the potential of machine learning in predicting archaeological probabilities and in in-terpreting the environmental influence on settlement patterns.