Abstract: In monitoring for uranium tailing, wireless sensor network (WSN) is superior to traditional line layout. Due to energy constraint, it is necessary to use energy efficiently in routing. After analyzing low energy adaptive clustering hierarchy (LEACH) and geographical adaptive fidelity (GAF) algorithms, an improved routing algorithm based on intersecting circle amd GAF is proposed. It uses location information of nodes and intersecting circle model to separate monitoring areas into a grid of circle. Cluster heads selection is carried out with surplus energy and distance to the virtual circle center. And, cluster heads choose the most optimal next hop to transmit information. The simulation results show that the algorithm can balance energy consumption effectively and extend network lifetime.