This paper presents a model which simulates changes in the airline system flight routing network under alternative policy scenarios. The model simulates a game between airlines, in which each airline increases flight frequency in order to maximize its own profit. The underlying modeling framework allows the relationships between changes in fares, passenger demand, infrastructure capacity constraints, flight delays, flight frequencies, and routing network to be simulated. The model is validated for a network of airports in the United States in 2005, before being applied to simulate changes in the same network through 2030 under two policy scenarios. Both scenarios limit airport capacity expansion: (i) in the whole system, and (ii) at Chicago O'Hare International, a primary hub airport, only. Simulated passenger demand, air traffic, flight delays, system CO2 emissions and Chicago O'Hare NOx emissions are compared to a baseline scenario in which airport capacity is expanded as planned by the FAA. Despite a significant impact on flight delays, the results show little impact of airport capacity constraints on system passenger demand, air traffic growth or CO2 emissions, but show a shift of connecting traffic away from congested hub airports at which capacity is limited to other less congested hub airports, thus reducing traffic growth at these congested airports, and reducing the growth in NOx emissions. Copyright © 2009 by University of Cambridge.