In this paper, an energy efficient data storage method for top-k query in wireless sensor networks is proposed. In the proposed method, the sensor network is divided into a number of grids and there is a grid head present at the center of each grid. Each grid head is assigned a definite sub-range of sensed data and the grid heads store only those sensed data within the assigned sub-range. Division of network into grids distributes the network traffic and reduces the traffic hotspot problem, which prolongs the network lifetime. The grids are assigned sub-ranges in a snake-like fashion so that minimum communication is required to query the top-k sensed data from the sensor network. We have proposed an energy efficient query algorithm to query the top-k value of sensed data. Experimental results show that our proposed method can prolong the lifetime of the sensor network as compared to the existing methods.