Can we trust proxy-based NAO index reconstructions?

Three existing cold season North Atlantic Oscillation (NAO) indices which w ere reconstructed from proxy data were assessed for their consistency and r obustness at the interannual and decadal timescales. All three indices were found to be inconsistent with an observed NAO index, whereas two other, ma inly instrumentally based indices, showed good agreement with observations. The combination of the proxy-based indices into one new index by means of multivariate regression yielded some improvement, such that the skill of th e new index was comparable or higher than the individual indices. However, the new index still did not correlate significantly with the instrumentally based reference time series prior to 1850. Our results suggest non-station arities in the statistical relationships between the proxy-based indices an d the NAO. This has important implications with regard to the choice of ver ification periods and predictor database for upcoming reconstruction attemp ts.