To support civil aviation carbon neutrality, this study develops an Improved Sparrow Search Algorithm-optimized BP neural network (ISSA-BP) model using 1996–2019 data. Five scenarios—Freeze, Baseline, Gradual Improvement, Substitution, and Technological Breakthrough—are simulated to forecast future CO? emissions. Results show the ISSA-BP model achieves a MAPE of 1.7382% and R2 of 0.9999, effectively avoiding local convergence. Simulations indicate that even under the optimal scenario, China’s aviation sector must reduce CO? emissions by 58%–62% to reach carbon neutrality. An integrated reduction pathway is proposed, including raising sustainable aviation fuel use to 70%, improving fuel efficiency, and enhancing carbon sinks. This study provides a scientific basis for aviation emission prediction and low-carbon strategy formulation.