Nonparametric estimation of functions and its applications

Reconstruction of the paleoclimate

Our long-standing collaboration with the paleoeclimate community has centered on using lake sediment fossil records to reconstruct past climate variation. In particular, statistical scale space methods have been developed for the quantification of credible climate features in different time scales in the presence of uncertainties caused by reconstruction and dating errors. Tutorials on the relevant scale space techniques, including the original inspiration, SiZer, and our Bayesian variant, BSiZer, summarize the main ideas. Important on-going work involves also Bayesian spatio-temporal temperature reconstructions using several sediment cores simultaneously.


BSiZer analysis of a diatom fossil based temperature reconstruction in northern Fennoscandia for the last 800 years. The upper panel: the reconstructed temperatures shown as dots together with three smooths obtained as posterior means. The lower panel: BSiZer map with the three smoothing levels shown using the same colors as in the upper panel. Red and blue indicate credibly warming and cooling climate, respectively.

Selected papers: