Transition towards carbon neutrality: Forecasting Hong Kong's buildings carbon footprint by 2050 using a machine learning approach

Abstract

Hong Kong has planned to achieve zero-carbon emissions before 2050. Hong Kong's building sector is responsible for approximately 60 % of carbon emissions, while its evolutionary trajectories and neutrality pathway remain unclear. To this end, we adopt the STIRPAT model and machine learning approach to identify the most relevant factors and determine their historical correlation with building energy consumption. Then, the out-of-sample prediction is conducted to explore the energy pathway and carbon footprint towards 2050 in the Hong Kong building sector under various assumptions and scenarios. The predicted results suggest that electricity consumption and carbon emissions will continue to increase following current trends. In a moderate scenario, if 2.2 % and 1.6 % of residential and commercial buildings are replaced annually with 2.75 % better energy-performance construction, their electricity consumption is likely to reach 30 %–40 % and 20 %–30 % from the 2015 level as the HK2050 required. With lower electricity emission intensity, the Hong Kong building sector could achieve near zero emissions. Our projection is valuable for developing energy-saving and emission-mitigation strategies towards various carbon neutrality goals in the building sector.