Bayesian Hierarchical Models for Climate Field Reconstruction; Palisades, New York, 8–11 February 2011; Inferring Earth's past climate from noisy, incomplete proxy time series remains an important challenge in the geosciences. Bayesian hierarchical models (BHMs), which have theoretical advantages over established regression methods based on empirical orthogonal functions (EOFs), are a powerful new method for spatially explicit climate field reconstruction. Whereas EOF-based approaches involve projecting the proxy and instrumental data sets onto reduced subspaces and then assuming a linear relationship between these transformed variables, BHMs allow for scientific knowledge about the target climate field to be included at the process level of the hierarchy and for each data type to have a different functional relationship with that target field. A workshop held at the Lamont-Doherty Earth Observatory of Columbia University provided in-depth exposure to BHMs for climate reconstructions to researchers who currently use EOF-based multivariate regression models for inferring past climate. In addition, the workshop explored how other established methods perform in relation to BHMs and examined the ability of each approach to characterize reconstruction uncertainties in practice.