夜间灯光数据与中国社会经济指标的关联研究

The Correlation Investigation between the Nighttime Light Data and Socio-Economic Data in China

摘要:

及时、客观的遥感卫星夜间灯光数据已被证明与人类的社会经济生活有着很强的相关性，获取夜间灯光数据与社会经济指数的相关性信息，对预测修正具有时间延迟、不准确问题的社会经济指数具有重大的意义。然而目前国内的这一方面的研究还处于极度缺乏的状态。因此本文以一线城市为例，研究了中国1994~2012年的DMSP/OLS稳定夜间灯光数据与国内生产总值GDP(Gross Domestic Product)和人口数的时间尺度和空间尺度上的相关性。此外，本文提出了系统化的DMSP/OLS夜间灯光数据的纠正方法，使得未经过星上检校的多时相夜间灯光数据具有了可比性和一致性。本文发现GDP和人口都与夜间灯光数据具有很强的相关性，且文中公布了夜灯与人口，GDP的相关性参数，为预测修正这些城市的社会经济数据提供帮助。

Abstract: Timely and reliable nighttime light data (NTL) derived by remote sensing satellite is closely related to social-economic life, and thus the study about the correlation information between NTL and socio-economic (e.g., population and gross domestic product (GDP)) is very crucial for predicting and revising the hysteretic and incorrect census data of socio-economy. However, at present, the related research in China is extremely rare. In view of this, in this paper, the DMSP- OLS Nighttime Lights (DMSP/OLS NTL) data of China between 1992 and 2012 were implemented to reveal the correlation information between NTL and socio-economic data from multiple perspectives (including space and time). Moreover, a systematic correction methods was developed to improve DMSP/OLS NTL data, which can remove the influence of water and gas flare, diminish blooming phenomenon, and make DMSP-OLS NTL data without onboard calibration become more continuous and comparable. The experiment in this paper revealed that the strong correlation between the socio-economic data (including GDP and population) and NTL. Furthermore, the detailed correlation and linear model was presented, which showed great potential for revising and predicting the census data of China.