Investigation of the relationship between the Pacific/North American (PNA) and North Atlantic oscillation (NAO) teleconnections, and Great Lake-effect snowfall

Large-scale mid-tropospheric air pressure variations from climatological means at a given location on Earth are suspected of exerting an influence on meteorological variables at the local scale many hundreds of miles distant; this type of relationship is known as a teleconnection. The hypothesis investigated here is whether, and to what degree, the Pacific/North American (PNA) and North Atlantic Oscillation (NAO) teleconnections may be found to be the driving factors behind anomalies in Great Lakes surface meteorological variables such as snowfall, precipitation, surface air temperature, wind trajectories, and ice cover. Special attention is placed on anomalous snowfall in the snowbelt regions of the Great Lakes.; The original contribution of this project to research in the field lies in the use of a new, high-resolution surface meterological dataset made possible The University of Georgia (Dyer 2005). This dataset, along with datasets provided by the National Centers for Environmental Prediction (NCEP) and the Climate Research Unit (CRU) of the University of East Anglia, Norwich, U.K. provide a heterogeneous collection of data for investigation of the teleconnection-surface meteorology question. Analyses include linear temporal trend analysis in each of the variables, followed by Pearson correlation coefficient and bivariate regression analysis between the PNA and NAO and surface meteorological variables.; The findings of this study indicate that, although neither the PNA nor NAO can account for the positive temporal trend noted in Great Lakes snowbelt snowfall, statistically significant relationships exist between each of the teleconnection indices and the surface meteorology of the Great Lakes region, and generally each index exerts an opposing influence. In particular, snowfall in the snowbelt of Lake Erie for the month of December is shown to be positively correlated (0.4) with the PNA index. The results further suggest that the Yarnal and Diaz (1986) formulation of the PNA index is a better predictor of snowfall anomalies in Great Lakes snowbelts than the Wallace and Gutzler (1981) version. Bivariate regression of anomalous snowfall against the Yarnal and Diaz PNA indicate the index can account for up to 30% of the December variance in the Lake Erie snowbelt.