Spatiotemporal Dynamics of Electricity Consumption in China

Nighttime light (NTL) data from the Defense Meteorological Satellite Program/Operational Linescan System (DMSP/OLS) provide information on nighttime luminosity, a correlation of built environment and energy consumption. This research intends to estimate spatial distribution of electricity consumption (EC) in mainland China, and analyze the temporal and spatial change of electricity consumption during 2000-2012. Nighttime light vegetation index (NVI), ratio nighttime light vegetation index (RNVI), difference nighttime light vegetation index (DNVI), normalized difference nighttime light vegetation index (NDNVI), soil adjusted nighttime light vegetation index (SANVI), and modified difference nighttime light vegetation index (MDNVI) were used to compensate for shortages in DMSP/OLS data. Moderate resolution imaging spectroradiometer (MODIS) NDVI products, China GIS database, and socioeconomic statistical data were also considered. An EC estimation model was used to obtain EC during 2000-2012. We divided EC into four ratings and analyzed spatiotemporal patterns using exploratory spatial data analysis tools (e.g., Moran's I and local indicators of spatial association-LISA statistics). Then we built a linear regression model of EC, and correlated with DMSP/OLS data to produce China's EC spatially. We used mean relative error (MRE) to compare our results and related research outcomes. Our result showed lower MRE, i.e., superior accuracy. EC grew quickly in China from 2000 to 2012 increasing from 6.79 to 14.82M kWh. Generating capacity and EC of 32 provinces, municipalities and autonomous regions have a strong spatial correlation. The proposed index combines information from DMSP/OLS NTL data and MODIS NDVI data for more detailed characterization of nighttime luminosity, and reduced NTL saturation. The index simplicity enables rapid characterization and monitoring of EC.