Abstract: Under the background of new urbanization, it is of great significance to study the low-carbon development for the high-quality development of China’s tourism industry. This study measures the tourism carbon emission with the method of “bottom-up”, and uses the panel threshold model in which new urbanization level is used as the threshold dependent variable to explore the stage characteristics and its heterogeneous characteristics of the impact of the new urbanization on tourism carbon emissions in the Yangtze River Economic Belt. The results show that :(1) There is a significant “N-shaped” curve relationship between the new urbanization and tourism carbon emission. (2) Under the guidance of new urbanization, per capita tourism consumption, tourism industrial structure and fixed capital stock of tourism enterprises have a significant restraining effect on tourism carbon emission. However，the level of regional economic development and technological development has become an important factor to promote the carbon emission of tourism industry. And environmental regulation is beneficial to regional tourism carbon emission reduction. (3) There is a significant interval effect between new urbanization and tourism carbon emissions. The tourism carbon emissions presents upward trend with the improvement of the new urbanization level when the new urbanization is lower than 0.319, while the improvement of the new urbanization level has a great inhibitory effect on the carbon emission of tourism industry when the new urbanization level is between 0.319 and 0.568, then the carbon emission of tourism will continue to increase with the improvement of the new urbanization level when the new urbanization level exceeds 0.568. Therefore, regional governments should formulate new urbanization development plans according to local conditions, and promote the tourism carbon emission reduction effect in the process of new urbanization from various aspects such as per capita tourism consumption level and tourism industry structure optimization.