Adaptive Multi-Scale Population Spatialization Model Constrained by Multiple Factors: A Case Study of Russia

Abstract

Population spatialization is the foundation for the visualization and analysis of population integrated with other information, such as environmental resources, economy, and public health. The existing population spatialization models have solved many problems for population distribution, but most of these studies have focused on a specific, single-scale approach and ignored the scale transformation for population spatialization. However, multi-scale visualization and the analysis of spatial information need multi-scale information. Meanwhile, the population distribution map as one kind of thematic map is always overlaid with the digital vector map or remote sensing map and visualized in the Web Geographic Information Systems (Web GIS), so it should adapt to the map scale showed in browser, when the user zoom in and zoom out. Hence adaptive multi-scale is necessary for population spatialization. Therefore, in this study, an adaptive multi-scale population spatialization model (APSM) is proposed with comprehensive factors constrained. These factors are residential area, land cover, public transport, hydrology, terrain, and climate. All of them are closely associated with population distribution. The overall methodology of APSM and the process for APSM are expounded in this paper. Meanwhile, with a case study of Russia, the processes of APSM for Russia are stated, and a population spatialization tool is implemented for expanding the application of APSM. The experimental analysis showed that APSM satisfied the requirement of geospatial analysis well and obtained a reliable accuracy.

Exploring resources and environmental carrying capacities at the county level: A case study of China’s Fengxian County

“…Next, we used standardized population data as dependent variable Y, and four types of land area data as independent variables X 1 , X 2 , X 3 , X 4, respectively. The intercept term in the linear regression model was set to zero [38]. According to the population dispersion coefficient (Table 4), we obtained the population of each 50 m × 50 m grid unit, as well the spatial discretization layout of population (Fig 2H).…”

Fine-Scale Spatiotemporal Analysis of Population Vulnerability to Earthquake Disasters: Theoretical Models and Application to Cities

“…To obtain data on the indicators in Table 1 at block scale, we needed to develop a population spatialization model. Different spatialization models of population data are used around the world, including multi-source information fusion [50], dasymetric mapping [51,52], and multiple regression methods [53], and many related models have been developed. Dong et al [54] comprehensively analyze and compare these models and identify their respective advantages and limitations.…”

From auxiliary data to research prospects, a review of gridded population mapping

“…Developing a highly automated platform for GPM may be a good approach. Although some GPM platforms (Hu et al, 2017; Lwin & Murayama, 2009; WorldPop, 2022; Yang et al, 2009) have been developed, they have various limitations, such as the inability to process raw auxiliary data, having only a single simple model and not having the ability to test the accuracy of the output data. We believe that an excellent GPM platform should meet the following four requirements: (1) The platform allows population and raw ancillary data to be downloaded online and processed in real‐time to obtain the data required for modeling; (2) The platform integrates multi‐model and multi‐scale (i.e., adjustable size of grid cells) GPM methods; (3) The platform uses a graphical interactive and visual interface, requiring little intervention and allowing non‐specialists to obtain target data quickly; and (4) The platform can test the accuracy of the output data.…”