A Method of Population Spatialization Considering Parametric Spatial Stationarity: Case Study of the Southwestern Area of China
Population is a crucial basis for the study of sociology, geography, environmental studies, and other disciplines; accurate estimates of population are of great significance for many countries. Many studies have developed population spatialization methods. However, little attention has been paid to...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-11-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/8/11/495 |
| _version_ | 1818272540574875648 |
|---|---|
| author | Junnan Xiong Kun Li Weiming Cheng Chongchong Ye Hao Zhang |
| author_facet | Junnan Xiong Kun Li Weiming Cheng Chongchong Ye Hao Zhang |
| author_sort | Junnan Xiong |
| collection | DOAJ |
| description | Population is a crucial basis for the study of sociology, geography, environmental studies, and other disciplines; accurate estimates of population are of great significance for many countries. Many studies have developed population spatialization methods. However, little attention has been paid to the differential treatment of the spatial stationarity and non-stationarity of variables. Based on a semi-parametric, geographically weighted regression model (s-GWR), this paper attempts to construct a novel, precise population spatialization method considering parametric stationarity to enhance spatialization accuracy; the southwestern area of China is used as the study area for comparison and validation. In this study, the night-time light and land use data were integrated as weighting factors to establish the population model; based on the analysis of variables characteristics, the method uses an s-GWR model to deal with the spatial stationarity of variables and reduce regional errors. Finally, the spatial distribution of the population (SSDP) of the study area in 2010 was obtained. When assessed against the traditional regression models, the model that considers parametric stationarity is more accurate than the models without it. Furthermore, the comparison with three commonly-used population grids reveals that the SSDP has a percentage error close to zero at the county level, while at the township level, the mean relative error of SSDP is 33.63%, and that is >15% better than other population grids. Thus, this study suggests that the proposed method can produce a more accurate population distribution. |
| first_indexed | 2024-12-12T21:43:42Z |
| format | Article |
| id | doaj.art-bc39f24f5b544d13aea81e4bd2caa084 |
| institution | Directory Open Access Journal |
| issn | 2220-9964 |
| language | English |
| last_indexed | 2024-12-12T21:43:42Z |
| publishDate | 2019-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj.art-bc39f24f5b544d13aea81e4bd2caa0842022-12-22T00:10:59ZengMDPI AGISPRS International Journal of Geo-Information2220-99642019-11-0181149510.3390/ijgi8110495ijgi8110495A Method of Population Spatialization Considering Parametric Spatial Stationarity: Case Study of the Southwestern Area of ChinaJunnan Xiong0Kun Li1Weiming Cheng2Chongchong Ye3Hao Zhang4School of Civil Engineering and Architecture, Southwest Petroleum University, Chengdu 610500, ChinaSchool of Civil Engineering and Architecture, Southwest Petroleum University, Chengdu 610500, ChinaState Key Laboratory of Resources and Environmental Information System, Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, ChinaSchool of Civil Engineering and Architecture, Southwest Petroleum University, Chengdu 610500, ChinaSchool of Civil Engineering and Architecture, Southwest Petroleum University, Chengdu 610500, ChinaPopulation is a crucial basis for the study of sociology, geography, environmental studies, and other disciplines; accurate estimates of population are of great significance for many countries. Many studies have developed population spatialization methods. However, little attention has been paid to the differential treatment of the spatial stationarity and non-stationarity of variables. Based on a semi-parametric, geographically weighted regression model (s-GWR), this paper attempts to construct a novel, precise population spatialization method considering parametric stationarity to enhance spatialization accuracy; the southwestern area of China is used as the study area for comparison and validation. In this study, the night-time light and land use data were integrated as weighting factors to establish the population model; based on the analysis of variables characteristics, the method uses an s-GWR model to deal with the spatial stationarity of variables and reduce regional errors. Finally, the spatial distribution of the population (SSDP) of the study area in 2010 was obtained. When assessed against the traditional regression models, the model that considers parametric stationarity is more accurate than the models without it. Furthermore, the comparison with three commonly-used population grids reveals that the SSDP has a percentage error close to zero at the county level, while at the township level, the mean relative error of SSDP is 33.63%, and that is >15% better than other population grids. Thus, this study suggests that the proposed method can produce a more accurate population distribution.https://www.mdpi.com/2220-9964/8/11/495population spatializationspatial stationaritygeographically weighted regressiondmsp/olsland use |
| spellingShingle | Junnan Xiong Kun Li Weiming Cheng Chongchong Ye Hao Zhang A Method of Population Spatialization Considering Parametric Spatial Stationarity: Case Study of the Southwestern Area of China ISPRS International Journal of Geo-Information population spatialization spatial stationarity geographically weighted regression dmsp/ols land use |
| title | A Method of Population Spatialization Considering Parametric Spatial Stationarity: Case Study of the Southwestern Area of China |
| title_full | A Method of Population Spatialization Considering Parametric Spatial Stationarity: Case Study of the Southwestern Area of China |
| title_fullStr | A Method of Population Spatialization Considering Parametric Spatial Stationarity: Case Study of the Southwestern Area of China |
| title_full_unstemmed | A Method of Population Spatialization Considering Parametric Spatial Stationarity: Case Study of the Southwestern Area of China |
| title_short | A Method of Population Spatialization Considering Parametric Spatial Stationarity: Case Study of the Southwestern Area of China |
| title_sort | method of population spatialization considering parametric spatial stationarity case study of the southwestern area of china |
| topic | population spatialization spatial stationarity geographically weighted regression dmsp/ols land use |
| url | https://www.mdpi.com/2220-9964/8/11/495 |
| work_keys_str_mv | AT junnanxiong amethodofpopulationspatializationconsideringparametricspatialstationaritycasestudyofthesouthwesternareaofchina AT kunli amethodofpopulationspatializationconsideringparametricspatialstationaritycasestudyofthesouthwesternareaofchina AT weimingcheng amethodofpopulationspatializationconsideringparametricspatialstationaritycasestudyofthesouthwesternareaofchina AT chongchongye amethodofpopulationspatializationconsideringparametricspatialstationaritycasestudyofthesouthwesternareaofchina AT haozhang amethodofpopulationspatializationconsideringparametricspatialstationaritycasestudyofthesouthwesternareaofchina AT junnanxiong methodofpopulationspatializationconsideringparametricspatialstationaritycasestudyofthesouthwesternareaofchina AT kunli methodofpopulationspatializationconsideringparametricspatialstationaritycasestudyofthesouthwesternareaofchina AT weimingcheng methodofpopulationspatializationconsideringparametricspatialstationaritycasestudyofthesouthwesternareaofchina AT chongchongye methodofpopulationspatializationconsideringparametricspatialstationaritycasestudyofthesouthwesternareaofchina AT haozhang methodofpopulationspatializationconsideringparametricspatialstationaritycasestudyofthesouthwesternareaofchina |