Mapping population on Tibetan Plateau by fusing VIIRS data and nighttime Tencent location-based services data. (June 2022)

• Record Type: Journal Article
• Authors: Ma, Xuankai; Yang, Zhaoping; Wang, Jingzhe; Han, Fang
• Is Part Of: Ecological indicators. Volume 139(2022)
• Publication Date: 2022-06
• Subjects: Population mapping -- Tibetan Plateau -- Nighttime Location-based Services Data -- VIIRS; Environmental monitoring -- Periodicals; Environmental management -- Periodicals; Environmental impact analysis -- Periodicals; Environmental risk assessment -- Periodicals; Sustainable development -- Periodicals; 333.71405
• DOI: 10.1016/j.ecolind.2022.108893
• Languages: English
• ISSNs: 1470-160X

Abstract

Graphical abstract: Highlights: For population modelling, remote sensing nighttime light data and nighttime LBS data were utilized. Population maps of the Tibetan Plateau at the city level were modelled by geographically weighted regression. The crowd exposure frequency characterized by nighttime LBS data contributed significantly to population modelling. The model predicted more accurate population maps than three mainstream international population datasets. The operational mechanism of the new demographic model was revealed. Abstract: Population mapping is one of the fundamental materials for regional sustainability studies. Most scholars applied remote sensing data with excessive indicators to fit the population distribution. Nevertheless, over-complex models were lack of accuracy. This paper proposed a population model in the Qinghai-Tibet Plateau as a study area; the model has consisted of Human Activity Extent and Crowd Exposure Frequency. Performed remote sensing land cover data, nighttime light data, and LBS geographic big data as candidate indicators for exploratory regression experiments eventually developed an optimal population model assembled by nighttime LBS data and nighttime light data. The model fits significantly better at the city level (R 2 values of 0.9922) and reduces the error compared with other studies and publicly available datasets (%RMSE values of 6.83%). For the first time, the model proposes that Crowd Exposure Frequency based on nighttime LBS data can provide effective population simulation in the global, nighttime light data gain compensation for it. Nighttime light data plays a dominant role in densely populated areas; at the same time, nighttime LBS revised its overestimation. They modified each other to make the model accuracy significantly elevated. The modelling framework allows dynamic and low-cost population estimates of ecologically vulnerable areas and thus serves sustainable regional development.