Very dense networks offer a better resolution of the physical world and therefore a better capability of detecting the occurrence of an event; this is of paramount importance for a number of industrial applications. However, the scale of such systems poses huge challenges in terms of interconnectivity and timely data processing. In this paper we will look at efficient scalable data acquisition methods for such densely instrumented cyber-physical systems. Previous research works have proposed approaches for obtaining an interpolation of sensor readings from different sensor nodes. Those approaches are based on dominance protocols, presenting therefore excellent scalability properties for dense instrumented systems. In this paper we propose an important advance to the state-of-the-art. Our novel approach not only incorporates a physical model to enable more accurate approximate interpolations but it also detects and self-adapts to changes in the physical model.