Comparative assessment of gridded population data sets for complex topography: a study of Southwest China
- PDF / 3,302,152 Bytes
- 19 Pages / 439.37 x 666.142 pts Page_size
- 17 Downloads / 193 Views
Comparative assessment of gridded population data sets for complex topography: a study of Southwest China Yongming Xu 1
& Hung
Chak Ho 2 & Anders Knudby 3 & Miao He 1
Accepted: 17 September 2020/ # Springer Nature B.V. 2020
Abstract Population estimates for high-resolution spatial grid cells data can reflect detailed spatial distribution of population, which are valuable for epidemiological studies, disaster risk assessments, and public resource allocation. However, choice of source data and methods for producing gridded population data sets can introduce spatial bias, especially in regions with complex geography. We assess the performance of four gridded population data sets from 2015 for the Dian-Gui-Qian region of Southwest China: Gridded Population of the World version 4 (GPW4), Global Human Settlement (GHS), LandScan, and WorldPop. At the town-scale, we found that GHS and WorldPop most closely resembled the 2015 population data used for validation. At the intra-town scale, for which spatially disaggregated population validation data do not exist, we compared each data set against Google Earth high-resolution images and found that WorldPop most closely resembled the population distribution that could be inferred from the imagery. We conclude that in general, WorldPop performs better than GPW, GHS, and LandScan. Keywords Gridded population data set . Assessment . Southwest China . GPW4 . GHS .
LandScan . WorldPop
Introduction Population is one of the key factors affecting social and economic development. Reliable population data can be used as a critical component in a wide range of public planning and
* Yongming Xu [email protected]; [email protected]
1
School of Remote Sensing and Geomatics Engineering, Nanjing University of Information Science & Technology, Nanjing 210044, China
2
Department of Urban Planning and Design, The University of Hong Kong, Pokfulam, Hong Kong
3
Department of Geography, Environment and Geomatics, University of Ottawa, Ottawa, Canada
Population and Environment
research, including epidemiological studies, health burden estimations, and disaster risk assessments (Fang et al. 2014; Linard et al. 2010; Ye et al. 2012; Snow et al. 2005; Calka et al. 2017; Leyk et al. 2019). Therefore, there are increasing requirements for high-quality population data in various disciplines (Gaughan et al. 2015; Freire et al. 2020). Though the need for population data with high spatial resolution is obvious, traditional population data obtained from statistical agencies provide only the total population in each spatial unit. However, the population across a district can exhibit high spatial variability (Briggs et al. 2007). Specifically, previous studies have noted that modifiable areal unit problem (MAUP) from scaling and zoning effects can be occurred when population data are aggregated into different levels of administrative units (Flowerdew 2011; Su et al. 2010). In addition, environmental and health studies often need to combine population data with other spatial data sets (e.g., land use,
Data Loading...
