Abstract: The purpose of this paper is aiming to build a low cost multicasting tree on a global coverage LEO satellite communication network.In this paper,the authors proposed a load balancing multicasting algorithm based on timescales for LEO satellite network(LBMT).LBMT collect the occupying state and calculate comparison cost of corresponding link and aiming to minimize the comparison cost rather than cost or delay of the tree.Several proposed multicasting algorithms aim for minimizing the delay or cost of multicast tree.However,the whole network traffic is not evenly distributed,multicast algorithms have been proposed usually deteriorate the network status under high load conditions because those algorithms often take the initiative to avoid the busy link.Simulation results shows that LBMT can decrease data loss rate under high traffic conditions with a little increase of transmission delay.Simulation results shows that LBMT can decrease data loss rate under high traffic conditions with a little increase of transmission delay.It can be concluded that the multicasting tree built by LBMT can efficiently avoid the busy ISLs and pick the idle ISLs above the developing region.The idle ISLs can be utilized by LBMT and the traffic of network is more balanced.