Population spatialization in Gansu Province based on RS and GIS

Spatialized population data is much closer to actual circumstances than population distribution represented in administrative units. It is also crucial for analyzing the relationship among economic growth, environment protection, resource utilization, and other interdisciplinary study. Remote Sensing (RS) and Geographic Information System (GIS) are both effective research tools of spatializing census data of human population. The advance in population spatial distribution during the last several decades based on RS and GIS was briefly introduced at the beginning. Through multi-variable regression analysis with statistical population data and fourteen land use indices in county scale for Gansu Province in 2000, it was found that there was a strong relation between population distribution at administrative unit level and land use knowledge. Thus it would be reasonable to generate the population grid from land use grids for Gansu Province in 2000. The final output was a population map at a 1 km2 resolution for study area. Corresponding validation was carried out at last.