To maximize the volume of data available for use on ground in near real time, many solutions have been proposed to perform data collection through satellite networks. Inspired by the backpressure algorithm, we are interested in the design of a network algorithm that not only exhibits long-term throughput optimality and but also delivers packets in reasonable end-to-end delays. Thus, we investigate the feasibility of backpressure algorithm on satellite networks and develop a throughput-optimal solution based on the utility maximize framework. For delay performance, we embed distance factor into the objective and derive a geographic-location-aware backpressure algorithm. The simulation results show that our algorithm is able to deliver packets in time and scale well as traffic load varies.