A good understanding of individual and collective wind farm operation is necessary for improving the overall performance of the wind farm “grid,” as well as estimating in real time the amount of energy that can be generated for effectively managing demand and supply over the smart grid. This paper proposes a scheme for compressing wind speed measurements exploiting both temporal and spatial correlation between the readings via distributed source coding. The proposed scheme relies on a correlation model based on true measurements. Two compression schemes are proposed, both of low encoding complexity, as well as a particle-filtering-based belief propagation decoder that adaptively estimates the nonstationary noise of the correlation model. Simulation results using realistic models show significant performance improvements compared to the scheme that does not dynamically refine correlation.