Name: Matthew Rogers
Title: The Pacific Decadal Precession and its Relationship to Tropical and Extratropical North Pacific Decadal Variability
Location: NWC 1350
Time: 03:00 PM
Series: Weather and Climate Systems
Abstract: Recent research uncovered evidence of a quasi-decadal mode of climate variability, termed the Pacific Decadal Precession (PDP), a counterclockwise progression of an atmospheric pressure anomaly dipole around the North Pacific at low frequencies (~10-20 years). In particular, the north-south oriented phase of the PDP, resembling the North Pacific Oscillation (NPO), is hypothesized to be partially maintained by Central Pacific sea-surface temperature anomalies, and thereby related to Central Pacific ENSO (CP-ENSO) events. Evidence for a relationship between CP-ENSO events and the north-south phase of the PDP would provide considerable insight for understanding the dynamics of the PDP. Indeed, accurate projections of the PDP can have significant implications for seasonal and interannual predictions of North American climate conditions such as drought and pluvial cycles in the Pacific Northwest, as it has been shown to be associated with decadal variability in seasonal precipitation over the Northwestern US.
This study first assesses the representation of the PDP in both reanalysis datasets and output from pre-industrial control runs of models in the Coupled Model Intercomparison Project Phase 6 (CMIP6) suite using time-extended empirical orthogonal function analysis, as outlined in previous literature. The relationship of the north-south phase of the PDP to the NPO, as well as potential linkages to the central tropical Pacific (i.e. CP-ENSO events), is then investigated through a combination of power spectrum analysis, point correlations, and empirical orthogonal function analysis. Findings indicate that low-frequency variability in the NPO is indeed closely related to the north-south phase of the PDP and that CP-ENSO events are indeed tied to variability in the north-south phase of the PDP. Applying these same methods to models in the CMIP6 suite achieves similar results, though the spatial structure and the evolution of the PDP varies considerably from model to model.