Estimating Socio-Economic Parameters via Machine Learning Methods Using Luojia1-01 Nighttime Light Remotely Sensed Images at Multiple Scales of China in 2018

Bin Guo,Yan Wang,Qiuji Chen,Pingping Luo,Yi Bian,Bo Zhang,Dingming Zhang,Yi Su,Xiaoxia Wang,Yarui Wu

doi:10.1109/access.2021.3059865

Abstract

Mapping socio-economic indicators with a raster format is still a great challenge. The nighttime light (NTL) datasets have been widely utilized to estimate the socio-economic parameters. However, the precision of the published datasets was too coarse to meet related issues such as flood losses assessment, urban planning, and epidemiological studies. The present study calibrated gross domestic product (GDP), population (POP), electric consumption (EC), and urban build-up area (B-A) at 100 m resolution for 45 cities of China in 2018 using Luojia1-01 NTL datasets via random forest (RF) as well as geographically weighted regression (GWR) model. The linear regression (LR), back propagation neural network (BPNN), and support vector machine (SVM) methods were selected for comparison with GWR and RF models. Besides, the Suomi National Polar-Orbiting Partnership-Visible Infrared Imaging Radiometer Suite (NPP-VIIRS) was chosen for comparison with Luojia1-01. The ten-folded cross-validation (CV) has been used for evaluating accuracy at county and city scales. Finally, the distribution maps of socio-economic parameters were illustrated and some findings were obtained. First, the validation results revealed that the calibration at the city-scale outperformed the county or district scale. Second, the precision of the Luojia1-01 NTL dataset surpassed the NPP-VIIRS NTL dataset on the same administrative scale except for some specific situations. Third, the precision of the simulation for the gross domestic product (GDP) is the highest than the others, followed by electric consumption (EC), build-up area (B-A), and population (POP). Fourth, the optimum model varied according to the socio-economic parameters. Fifth, the distribution of socio-economic parameters exhibited obvious spatial heterogeneity. This paper can supply scientific support for calibrating socio-economic parameters in other regions.

Highlights

Socio-economic parameters are valuable data sources for governments making decisions and scientific researches [1], [2]
The objectives of this study are: (i) to calibrate the socio-economic parameters including gross domestic product (GDP), population (POP), electricity consumption (EC), and urban build-up area (B-A) based on linear regression (LR), geographic weighted regression (GWR), Back Propagation (BP) neural network, support vector machine (SVM), and Random Forest (RF) using statistical, NPP-Visible Infrared Imaging Radiometer Suite (VIIRS), and Luojia1-01 data of China in 2018, (ii) to validate and compare the simulation results determined by different models and multi-source data at multiple scales, (iii) to evaluate the feasibility of calibration methods using nighttime light (NTL) data to simulate socio-economic parameters
For the same socio-economic parameter, the accuracy of the simulation is better at the city scale with relatively higher correlation coefficients (R2) and lower root mean square error (RMSE) as well as mean absolute error (MAE) than the county or district scale based on the same NTL dataset excluding some special situations

Summary

Introduction

Socio-economic parameters are valuable data sources for governments making decisions and scientific researches [1], [2]. The world’s second-largest economy, is undergoing massive infrastructure construction [3], [4]. Industrialization, urbanization [5] and human. The traditional statistical datasets with the political division scale are inadequate in describing socio-economic phenomena because the frequency of updates for statistical data is always yearly [6]. The spatial heterogeneity of socioeconomic phenomena within the administrative division can hardly be described for the data accuracy is restricted [7]– [12]. The classic investigation methods for statistical data are time and finance-consuming [13].

Objectives

Methods

Results

Discussion

Conclusion