<p>Rapid motorization poses challenges to transport decarbonization and air quality in China. The built environment shapes travel behavior, and transit-oriented development (TOD) has emerged as a key strategy for low-carbon mobility. This study develops a predictive model that links built environment indicators to travel modal shares and associated emissions. The model identifies the effects of 14 indicators on six travel modes: private cars, buses, metros, electric bikes, shared bikes, and walking, and simulates a level of integration (<i>LOI</i>) to represent the built environment attractiveness for different travel modes. Based on the <i>LOI</i>, travel modal shares are predicted and translated into carbon and air pollutant emissions using a bottom-up approach. The model is applied to three metro-centered precincts with distinct urban forms in Nanjing. The results reveal substantial differences in travel emissions across urban forms: in the car-dependent suburban precinct, the private car modal share reaches 38.7%, while in the transit-rich historic core it is 15.0%, leading to per capita annual carbon dioxide (CO₂) emissions of 857 kg and 429 kg, respectively. Most air pollutant emissions follow trends similar to CO₂, except for sulfur dioxide (SO₂), which exhibits higher per capita emissions in the historic core due to larger shares of electricity-based modes. A TOD retrofitting scenario for the suburban precinct indicates that targeted interventions can reduce travel CO₂ emissions by 6.1%, and air pollutant emissions by 2.8%–6.4%. Overall, the proposed model provides a tool to assess travel-related emissions across contrasting urban forms and identify TOD pathways for low-carbon travel and improved air quality.</p>