Assessment of Population Density in Southern Siberia Using Remote Sensing Data

Abstract:

The detailed information on the spatial distribution of the population density is essential for understanding urban dynamics and for analyzing various social-economic, environmental and political factors. Population migration processes along with the climate change are particularly relevant currently and in some cases are related. For the territory of Siberia the population density is significantly correlated with the spatial distribution of climate severity. For instance, on the territory of the Krasnoyarsk krai there are both areas with relatively comfortable climatic conditions (forest-steppe and steppe zones) and extremely uncomfortable regions in the permafrost zone. By the end of this century, in accordance with the projected climatic changes, a fundamental redistribution of the entire ecological-resource potential and the capacity of ecological niches that are significant for humans will occur.

In recent decades, remote sensing techniques have often been used to estimate population density, especially using night time light (NTL) data. In our study we used a several remotely sensed datasets to create evaluate the spatial distribution of population in the southern and central regions of the Krasnoyarsk krai. We used the National Polar-Orbiting Partnership/Visible Infrared Imaging Radiometer Suit (NPP/VIIRS) day/night band as a source of NTL data. We also used NPP/VIIRS normalized difference vegetation index (NDVI) measurements to improve the estimation of population density.

Using the DNB and NDVI data for the year of 2012 we estimated the spatial distribution of remotely sensed metrics with the spatial resolution of 500 m and compared it with the census data of 2010. Least square regression model was used to quantitatively characterize the relationship between the population density according to census data and the average values of remotely sensed metrics. The results indicated that the estimated population density was in fairly good correspondence with the census data.

Acknowledgments: This research was supported by the Russian Foundation for Basic Research (RFBR), project 19-45-240004.