Integrating carbon emissions and sinks for a just transition: A spatial framework for the Guangdong-Hong Kong-Macao Greater Bay Area

Achieving carbon neutrality in megacity regions demands spatially differentiated strategies that coordinate emission reductions, carbon sink enhancement, and social fairness. This study provides a spatially integrated, data-driven framework that unifies grid-level carbon emissions and carbon sinks with machine learning–based scenario forecasting. It overcomes the coarse spatial resolution and sectoral limitations of traditional models. Developed for the Guangdong-Hong Kong-Macao Greater Bay Area (GBA), this approach integrates high-resolution local climate zone (LCZ) mapping, socioeconomic data, sectoral energy statistics, and direct carbon sink measurements to assess four future policy scenarios through 2060. Unlike previous models, this framework couples supply- and demand-side interventions with spatially explicit, population-weighted equity metrics, which enables precise identification of emission hotspots, social risk zones, and mitigation opportunities. Scenario-specific machine learning models compare a business-as-usual (BAU) trajectory to three carbon neutrality (CN) pathways. The integrated CN scenario yields the greatest emission cuts, lowering annual carbon emissions to 41.95 million tons by 2060, far below the 399.26 million tons projected under BAU. Spatial results indicate that densification, industrial upgrades, and urban greening reduce both emissions and social inequalities, while the period from 2040 to 2050 is identified as the key window for implementing advanced retrofits and carbon capture. It presents a phased roadmap: end new coal and expand renewables in the 2020s, increase city greening and electric transport in the 2030s, implement carbon capture in the 2040s, and require net-zero and circular economy by 2060. This framework offers practical and data-driven guidance for urban regions pursuing deep and fair decarbonization.