Climate dynamics

Abstract:

We employ multi-ensemble Met Office Decadal Prediction System hindcasts to analyse the impact of atmospheric winds and North Atlantic Oscillation (NAO) phases on the overturning circulation in the North Atlantic Ocean. A positive NAO phase is generally associated with an anomalously strong and/or northward shifted jet stream in the North Atlantic, and the vice-versa is true for a negative NAO phase. As a consequence of relatively strong winds, oceans tend to lose more heat to the atmosphere in winter in many parts of the subpolar North Atlantic Ocean. This process is expected to create negative anomalies in sea surface temperature and generate more dense water on the ocean surface at high latitudes resulting in a strengthening in the overturning circulation. Here, we examine the sensitivity of the overturning circulation to NAO phases in multi-ensemble decadal hindcasts to understand how the interior ocean responds to different NAO phases. For this purpose, we analyse the changes in east-west density contrasts, upper ocean heat content, mixed-layer depth, meridional heat and salt transport in different oceanic regions, i.e. Labrador Sea, Irminger Sea and Nordic Seas. In particular, we perform a linear regression analysis for the above-mentioned diagnostics and NAO indices to assess how sensitive the upper ocean is to changes in the atmospheric state. We further compare our results against reanalysis data and in-situ observations.

Abstract:

The leading mode of wintertime atmospheric variability over the North Atlantic-North Eurasia sector is dominated by the North Atlantic Oscillation (NAO) and accounts for more than one third of the total variability. This study explores the influences of the leading mode on decadal climate variability of Northern Eurasia. We focus on the little-explored decadal covariations of surface air temperature (SAT), snowfall, snow water equivalent (SWE) and snow cover over the region, using extensive model output from the Coupled Model Intercomparison Project sixth phase. Recent decadal trends (−0.92σ per decade) in the leading mode identified, are found to be largely a manifestation of internal climate variability (at least two thirds from the most conservative estimate). These internally-generated decadal trends strongly contributed to recent trends in SAT, snowfall, SWE and snow cover over Eurasia. External forcings should have played a minor role over Eurasia as they usually suggest opposite decadal trends to those observed. An exception is found for snowfall and SWE in east Eurasia, for which external forcings may have driven a large part of the recent upward trends, equally as important as the NAO-dominated mode. This points to a complex interplay between internally-generated and externally-forced climate variability over Northern Eurasia. Model discrepancies are identified in reproducing the linkages between the leading mode and the Eurasian surface climate variability. The internally-generated variability of this leading mode thus represents a large source of uncertainty in future decadal climate projections over Eurasia and, due to the memory effects of snow, also in modelling springtime climate variability.