Integrating local climate zones and spatial modeling for carbon emission forecasting in the Guangdong-Hong Kong-Macao Greater Bay Area toward 2060

Abstract

High-resolution and sector-specific spatial prediction of carbon emissions is essential for developing effective urban planning strategies to mitigate climate change. This study introduces an innovative approach by integrating the Local Climate Zone (LCZ) scheme and calculating landscape metrics indices as impact factors to enhance the spatial precision of carbon emission predictions. Using the Long-range Energy Alternatives Planning System (LEAP) model, enriched with machine learning and sector-specific analysis, this research predicts spatial carbon emissions in the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) from 2020 to 2060 at a fine resolution of 500 m × 500 m under the Business-As-Usual scenario. Results indicate a peak in emissions around 2030, followed by a targeted 22.7 % reduction by 2060 compared to 2020 levels. While a shift from coal to cleaner energy sources is evident, the increasing dependence on natural gas raises concerns. The study highlights that urban morphology, population density, and LCZ classifications significantly shape emission pathways. Quantitative modeling reveals that morphological features such as LCZ-based aggregation and connectivity indices have measurable effects on emissions across sectors. The findings emphasize the need for integrating spatial planning with energy policies and provide a replicable framework for metropolitan regions, which could guide dynamic policy strategies for urban development.