Interactive geographical and temporal weighted regression to explore spatio-temporal characteristics and drivers of carbon emissions
Wei Tu, Congjun Rao, Xinping Xiao, Fuyan Hu, Mark Goh
Abstract
Countries need a science-informed strategy to manage carbon peaking and carbon neutrality. This study extends the geographically and temporally weighted regression (GTWR) model to include the GeoDetector's factor interaction detection plate to investigate the spatio-temporal characteristics of the factors influencing regional carbon emissions in the Yangtze River Economic Belt (YEB), an important economic area in China. The results from the proposed interactive geographically and temporally weighted regression (IGTWR) model indicate that the evolution of carbon emissions can be categorized into two phases in the temporal dimension. In terms of spatial distribution, the carbon emissions of the YEB are distributed in a northeast southwest direction, are centered in Hubei Province and cover a broad geographical range. Both the drivers of carbon emissions and their factor interactions possess spatial heterogeneity. • Explore the spatio-temporal characteristics and driving factors of carbon emissions. • Propose a new interactive geographically and temporally weighted regression model. • Interactions of factors show dual-factor strengthening and nonlinear enhancement. • Both single and interaction factors show spatial heterogeneity on carbon emissions.