Study on Spatiotemporal Coupling Between Urban Form and Carbon Footprint from the Perspective of Color Nighttime Light Remote Sensing

Abstract

Highlights: This study develops an innovative color nighttime light remote sensing approach and applies it to systematically quantify the spatiotemporal coupling relationship and underlying mechanisms between urban form and carbon emissions in a major Chinese metropolis. What are the main findings? A high-resolution color nighttime light remote sensing imagery (color-NLRSI) dataset was generated using an NSCT-IHS dual-transform fusion method, significantly improving the accuracy of built-up area extraction and spatial delineation compared to traditional NPP-VIIRS data. Applied to Guangzhou, this method revealed that urban spatial expansion is the dominant influencing factor of carbon emissions, with a standardized influence coefficient of 22.43%, substantially higher than those of economic development (10.34%) and urbanization rate (14.91%). What is the implication of the main finding? The color-NLRSI dataset and the integrated technical framework provide a reliable, high-resolution foundation for accurately monitoring carbon emissions and urban dynamics, enabling fine-grained environmental governance and low-carbon spatial planning. The integration of color nighttime light data with spatiotemporal modeling reveals a distinct “core-periphery” heterogeneity in urban development-carbon emission relationships, highlighting the need for region-specific policies to mitigate carbon lock-in and promote sustainable transformation. This study addresses the limitations of traditional nighttime light remote sensing data in ground object feature recognition and carbon emission monitoring by proposing a fusion framework based on Nonsubsampled Contourlet Transform (NSCT) and Intensity-Hue-Saturation (IHS). This framework successfully generates a high-resolution color nighttime light remote sensing imagery (color-NLRSI) dataset. Focusing on Guangzhou, an important city in the Guangdong-Hong Kong-Macao Greater Bay Area, the study systematically analyzes the spatiotemporal coupling mechanism between urban form evolution and carbon footprint by integrating multiple remote sensing data sources and socio-economic statistical information. Key findings include: (i) The color-NLRSI dataset outperforms traditional NPP-VIIRS data in built-up area extraction, providing more accurate spatial information by refining urban boundary recognition logic. (ii) Spatial correlation analysis reveals a remarkably strong positive relationship between built-up area expansion and carbon emissions, with the correlation coefficient for numerous districts exceeding 0.9. High-density built-up areas are strongly associated with a carbon lock-in effect, hindering low-carbon transformation efficiency. (iii) Geographically Weighted Regression analysis demonstrates that in population-polarized regions, the impact coefficient of built-up area expansion on carbon emissions is notably high at 0.961. This factor’s association (22.43%) surpasses economic development (10.34%) and urbanization rate (14.91%). The established “data fusion—dynamic monitoring—mechanism analysis” technical system, which generates a novel high-resolution color-NLRSI dataset and reveals a distinct ‘core-periphery’ heterogeneity pattern in Guangzhou, demonstrating that urban expansion is the dominant driver of carbon emissions. This approach offers a scientific basis for tailored urban low-carbon development strategies, spatial optimization, and enhanced precision in carbon emission monitoring.