Understanding Spatial Patterns And Driving Mechanisms Of Ancient Chinese Population Based On Macroecology Theory

Understanding human-environment interactions at both broad spatial and temporal scales is a central problem of human macroecology. Many studies have focused on impacts of humans on the environment, whereas few have explored the influence of the environment on humans. In this study, we aim to analyze human abundance spatial patterns’ drivers involving climate, topography, dispersal and agricultural resources and shed light on the underlying potential processes by applying non-spatial and spatial statistic models to the reconstructed human population density patterns in ancient agricultural China, and therefore contribute to better understanding of humans’ adaptation to varying environment and provide theoretical basis for the predictions about humans’ responses to global environmental change.We studied the populations with complete official records in ancient Chinese dynasties (from Xihan to Qing), mainly covering North China, East China, South China and Southwest China. The database spanned-1800 years in time,30° in longitude,20° in latitude, and 4.5 km in elevation. Our municipal population data and maps were derived from bibliographies on historical demography of China and atlas of Chinese dynasties. Population data, environmental variables and dispersal data were calculated using grain size of 100 km×100 km. We applied both ordinary least squares (OLS) and simultaneous autoregressive (SAR) models to fit the population data with explanatory variables. The pseudo-coefficients of determinant of the trend components of SARs represented the explanatory powers of the non-spatial trends of explanatory variables. Our study showed that agricultural population in a stable social environment was prone to aggregate in plain areas, whereas under turbulent conditions population tended to distribute over more or less mountainous regions. Convenience of transportation to and from the capital was the most powerful determinant among the four sets of explanatory variables in shaping human population density patterns. In most cases, human population density could be fitted with normal curves along both annual and quarterly precipitation gradients, confirming the niche-based hypothesis for the general distribution-abundance pattern. Curves of human population density against annual mean temperature also represented normal distribution until the 12th century. Natural resources were more responsible for human population patterns under turbulent conditions than under stable ones.