Quantifying embedded carbon emissions in China's light industry exports using a global multi-regional input-output analytical approach

China's light industry is very important to global value chains, but we still don't know enough about how it contributes to embodied carbon emissions at the sector level. This research utilizes a global multi-regional input-output framework derived from the World Input-Output Database to measure both production-based and consumption-based embodied carbon emissions linked to China's light industrial exports from 2002 to 2016. The analysis uses the Leontief inverse matrix and sectoral carbon intensity coefficients to follow direct and indirect carbon transfers between 42 economies and 56 sectors. The results show that China is always a net exporter of embodied carbon emissions, especially in carbon-heavy industries like textiles, leather, and paper products. Production-based accounting assigns considerable emissions to domestic industrial activities, whereas consumption-based analysis reveals that a big portion of these emissions is influenced by international demand. This difference shows that there are structural imbalances in global carbon responsibilities and provides real-world proof of how carbon leakage works in international trade. To bolster robustness, the work employs cross-database validation with EXIOBASE and EORA, thereby affirming that WIOD-based estimations maintain consistency within acceptable analytical margins. The study connects embodied carbon flows to carbon leakage, comparative advantage in pollution-heavy industries, and environmental justice issues. This adds to the continuing discussions about how to make trade more sustainable and fair climate governance. The findings provide pertinent insights for sector-specific decarbonization, preparedness for carbon border adjustments, and synchronized global mitigation efforts amid the developing climate policy frameworks.