Identifying Dynamically Induced Variability in Glacier Mass-Balance Records

Glacier mass balance provides a direct indicator of a glacier's relationship with local climate, but internally generated variability in atmospheric circulation adds a significant degree of noise to mass-balance time series, making it difficult to correctly identify and interpret trends. This study applies "dynamical adjustment'' to seasonal mass-balance records to identify and remove the component of variance in these time series that is associated with large-scale circulation fluctuations (dynamical adjustment refers here to a statistical method and not a glacier's dynamical response to climate). Mass-balance records are investigated for three glaciers: Wolverine and Gulkana in Alaska and South Cascade in Washington. North Pacific sea level pressure and sea surface temperature fields perform comparably as predictors, each explaining 50%-60% of variance in winter balance and 25%-35% in summer balance for South Cascade and Wolverine Glaciers. Gulkana Glacier, located farther inland, is less closely linked to North Pacific climate variability, with the predictors explaining roughly 30% of variance in winter and summer balance. To investigate the degree to which this variability affects trends, adjusted mass-balance time series are compared to those in the raw data, with common results for all three glaciers; winter balance trends are not significant initially and do not gain robust significance after adjustment despite the large amount of circulation-related variability. However, the raw summer balance data have statistically significant negative trends that remain after dynamical adjustment. This indicates that these trends of increasing ablation in recent decades are not due to circulation anomalies and are consistent with anthropogenic warming.