Northern High Latitude Climate Variability of the Last Millenium

This work explores the causes of northern high-latitude climate variations over the last millennium, and industrial and future periods. Attribution studies are performed on a suite of global climate simulations, and four historical reconstructions of Greenland surface temperatures and precipitation (two of which are new to this work). The simulations followed the protocols of the Palaeoclimate Modelling Intercomparison Project 3 and Coupled Model Intercomparison Project 5. At least half of the multi-decadal variability in simulated Greenland climate variations over the last millennium is reproduced by a linear, empirically-generated model including terms for volcanic emissions, solar insolation changes (including total solar irradiance and orbital components) and an index associated with latitudinal shifts in the North Atlantic jet. Empirical model parameters are obtained by regressing simulated Greenland temperatures and precipitation against time series for each of the response variables. Greenhouse gas radiative forcing changes are unimportant to simulated Greenland conditions over the last millennium, although they dominate after the mid-20th century. Most of the historical Greenland climate reconstructions are restricted to the industrial period, due to a lack of spatially-comprehensive climate records. They exhibit substantial differences in the timing, phasing and amplitudes of past climate variations, due to regional sensitivities in the source data and the reconstruction methodologies. Reconstructions indicate that Greenland temperatures did not begin to follow hemispheric greenhouse gas warming patterns until the mid-1990s. This discrepancy indicates either that the warming hiatus was associated with internal climate variability, or that the simulations are missing processes important to Greenland climate. For example, indirect effects of anthropogenic aerosols are not captured in the climate model employed here. All of the external climate forcings included in the empirical models initiate significant simulated climate responses in a number of northern hemispheric regions over the last millennium. These forcing responses are not independent, however, except in Tropical regions. Particularly in the Arctic and North Atlantic, nonlinear interactions between the forcings work together to generate the transition between the Medieval Climate Anomaly and the Little Ice Age. Internal variability also plays an important role, even on hemispheric averages.Ph.D