In this paper, the spatio-temporal saliency is applied to weather radar observations with the goal of exploiting information-theory-based methods for adaptive weather sensing with Phased Array Radars (PAR). PARs can dynamically and adaptively change the beam pointing direction and associated acquisition parameters, which can potentially lead to improved warnings and forecasts. Using spatio-temporal saliency, scanning strategies can be tailored to extract maximum information from the atmosphere. In this paper we present the spatio-temporal saliency as a driver for adaptive sensing, its behavior for different types of weather phenomena, and its applications to adaptive weather sensing. We demonstrate the advantages of this algorithm using data collected by Next Generation Weather Radars (NEXRAD).