The importance of stratospheric initial conditions for winter North Atlantic Oscillation predictability and implications for the signal‐to‐noise paradox

Abstract

This study investigates the influence of atmospheric initial conditions on winter seasonal forecasts of the North Atlantic Oscillation (NAO). Hindcast (or reforecast) experiments -which differ only in their initial conditions -are performed over the period 1960-2009, using prescribed sea surface temperature (SST) and sea-ice boundary conditions. The first experiment ("ERA-40/Int IC") is initialized using the ERA-40 and ERA-Interim reanalysis datasets, which assimilate upper-air, satellite and surface observations; the second experiment ("ERA-20C IC") is initialized using the ERA-20C reanalysis dataset, which assimilates only surface observations. The ensemble mean NAO skill is largest in ERA-40/Int IC (r = 0.54), which is initialized with the superior reanalysis data. Moreover, ERA-20C IC did not exhibit significantly more NAO hindcast skill (r = 0.38) than in a third experiment, which was initialized with incorrect (shuffled) initial conditions. The ERA-40/Interim and ERA-20C initial conditions differ substantially in the tropical stratosphere, where the quasi-biennial oscillation (QBO) of zonal winds is not present in ERA-20C. The QBO hindcasts are highly skilful in ERA-40/Int IC -albeit with a somewhat weaker equatorial zonal wind amplitude in the lower stratosphere -but are incorrect in ERA-20C IC, indicating that the QBO is responsible for the additional NAO hindcast skill; this is despite the model exhibiting a relatively weak teleconnection between the QBO and NAO. The influence of the QBO is further demonstrated by regressing out the QBO influence from each of the hindcast experiments, after which the difference in NAO hindcast skill between the experiments is negligible. Whilst ERA-40/Int IC demonstrates a more skilful NAO hindcast, it appears to have a relatively weak predictable signal; this is the so-called "signal-to-noise paradox" identified in previous studies. Diagnostically amplifying the (weak) QBO-NAO teleconnection increases the ensemble-mean NAO signal with negligible impact on the NAO hindcast skill, after which the signal-to-noise problem seemingly disappears.

How confident are predictability estimates of the winter North Atlantic Oscillation?

“…We have thus far emphasised that decadal variability and insufficient sample size can affect the robustness of skill and confidence estimates of hindcasts that are only a few decades long. However, it is also possible that hindcasts can display under‐confidence in the RPC measure whilst seemingly having the correct RMSE/spread relationship; such a situation can occur through the model having a weak predictable signal (see also O'Reilly et al, ). This can be demonstrated using the ASF‐20C NAO hindcast, which actually has RPC ∼ 1.…”

“…The focus of this article is the predictability of the interannual variability of the winter North Atlantic Oscillation (NAO), the main mode of variability over the Euro‐Atlantic region on a range of time‐scales from days to seasons and longer. Teleconnections from ENSO, precursor temperatures of the North Atlantic Ocean, Arctic sea ice over the Barents‐Kara Sea and the Quasi‐Biennial Oscillation in the stratosphere are proposed physical mechanisms that can contribute to predictive NAO skill (Eade et al, ; Scaife et al, ; Dunstone et al, ; Hansen et al, ; Wang et al, ; O'Reilly et al, ). Indeed, significant ensemble mean correlation skill of approximately r ≈ 0.6 for the NAO (Scaife et al, ) and its hemispheric counterpart of the Arctic Oscillation (AO: Derome et al, ; Stockdale et al, ; Kumar and Chen, ) were reported for some dynamical forecasting systems (see also Baker et al, ).…”

Jet latitude regimes and the predictability of the North Atlantic Oscillation

“…A good overview of these can be found in Smith et al (2016). Some of the more consistently proposed drivers include Atlantic sea surface temperatures (Czaja and Frankignoul 1999;Rodwell et al 1999;Saunders and Qian 2002;Dunstone et al 2016), the El Niño Southern Oscillation (ENSO) (Dong et al 2000;Moron and Gouirand 2003;Toniazzo and Scaife 2006;Brönnimann 2007;Li and Lau 2012;Drouard et al 2013;Jiménez-Esteve and Domeisen 2018) the stratospheric polar vortex (Baldwin and Dunkerton 2001;Dunstone et al 2016;Kidston et al 2015), the quasi biennial oscillation (QBO) (Anstey et al 2013;Scaife et al 2014b;O'Reilly et al 2019) and Arctic sea ice, particularly in the Kara region (Strong and Magnusdottir 2011;Koenigk et al 2016;Wang et al 2017).…”

When and where do ECMWF seasonal forecast systems exhibit anomalously low signal‐to‐noise ratio?

“…There are a number of possible reasons why the ECMWF forecast systems might be unable to capture the available skill in early winter. O'Reilly et al (2018) showed evidence that the link between the QBO and the NAO was too weak in a version of the ECMWF model. QBO effects in the polar stratosphere, while often hard to untangle from other drivers of variability, are thought to be most significant from October to December (Garfinkel and Hartmann, 2007).…”