A load balancing-routing wavelength assignment (LBRWA) algorithm based on user traffic distribution for satellite optical networks (OSNs) is proposed. The ant colony algorithm is employed to find the optical path so that the OSNs can achieve effective load balancing. The LBRWA optimization model in the OSNs based on global user traffic distribution is established. It searches for the next-hop link according to the link duration and wavelength idle rate and introduces random disturbance to avoid local optimum. The statistical average optical path cost corrected by the user traffic intensity is employed to update the link status; the traffic is directed to the satellite nodes in the non-hot spot area to find the load balancing optimal optical path for user requests. Numerical simulation results show that compared with traditional elastic load balancing(ELB) algorithm based on link-state, LBRWA algorithm reduces the congestion rate by 20.49%. In addition, LBRWA has better performance in terms of resource utilization, traffic distribution index, and max normalized inter-satellite link load, thereby effectively realizing load balancing in OSNs.