Urban climate research has surged in the past decades due to its unique position in addressing the concomitant issues of the rapid global urbanization and emergent climate changes. Nevertheless, there is still a lack of holistic and system-based toolkit for investigating complex dynamics of urban climate systems, especially those of hydroclimate extremes. Here we propose a novel framework based on complex networks for analyzing the topological structure of the urban climate in the contiguous United States (CONUS). In addition, we simulate CONUS urban networks using Kuramoto model and the master stability function formalism, which sheds new light on the occurrence and evolution of heat waves as dynamic synchronization processes. It is found that the CONUS urban climate networks are highly modular with spatial hierarchical (hub-periphery) organization. The small-world effect is also manifest in these networks with the characteristic path length between any pair of nodes no more than 5. For globally connected urban networks, as the sparsity of connections increases (with threshold of correlation coefficient), the path length decreases roughly linearly, whereas the synchronizability reduces exponentially.