Extreme temperature regimes (ETRs) such as cold air outbreaks (CAOs) and warm waves affect regional economies and human safety over large portions of the United States via their significant impacts on energy consumption, local agriculture, and human health. Regional CAO frequency and intensity are strongly modulated by larger-scale low frequency modes of variability. Since ETR behavior is modulated by large-scale modes of natural variability, two key scientific issues arise in relation to the numerical simulation of ETR events: First, the ability of climate models to accurately represent ETR behavior will be limited by the accuracy of the models used in representing the primary modes of low frequency variability. Second, future changes in ETR behavior will be strongly linked to the evolving characteristics of low frequency modes. Our proposed research will study the long-term variability of ETR events in relation to the non-stationary behavior of relevant modes of low frequency variability. Our primary scientific goals and objectives are: 1) Better quantify the role of natural modes of low frequency variability in determining ETR behavior 2) Diagnose the limitations of climate models in representing ETR-low frequency mode linkages 3) Assess likely future changes in ETR behavior and ETR-low frequency mode linkages We will pursue our study via statistical and dynamical diagnostic analyses of both observational reanalyses and long-term coupled model simulations. We will first construct a synoptic-dynamic climatology of ETR for the boreal cool season. Multivariate statistical methods will then be applied to assess regional ETR modulation by prominent low frequency modes such as ENSO, PDO, PNA, NAO and AO/NAM. We will further examine the behavior and low frequency modulation of ETRs in coupled climate model simulations, employing simulations related to the Coupled Model Intercomparison Project Phase 5 (CMIP5). We will first assess the fidelity with which ETRs and ETR-low frequency mode linkages are represented in the pre-industrial and historical experiments of CMIP5. We will additionally diagnose model limitations in representing the fundamental characteristics of the critical low frequency modes, themselves. Simulations from multiple future climate projection experiments of CMIP5 will then be studied to assess how the regional characteristics of ETR events are likely to vary in association with changes in low frequency modes under global warming. We hypothesize that future ETR trends over North America will be jointly controlled by (natural and forced) decadal variability in the behavior of PNA and NAM. Our analyses will test this idea while serving to better constrain likely future decadal changes in the characteristics of ETRs, PNA and NAM. Our proposed research will provide important new information about the modulation of extreme events by low frequency modes of natural variability, how this modulation is likely to change in coming decades and what are the likely impacts of this change on weather extremes such as ETRs.