Comparison of Subseasonal‐to‐Seasonal Model Forecasts for Major Stratospheric Sudden Warmings

Abstract

This study explores forecasts for five distinctive major stratospheric sudden warmings (MSSWs) using subseasonal‐to‐seasonal prediction data with lead times up to about 2 weeks. Results reveal model‐to‐model variability of the forecasts, as some models, such as the European Centre for Medium‐Range Weather Forecasts and National Centers for Environmental Prediction models, forecast the MSSWs better. Results also demonstrate greater difficulty of forecasting the vortex split MSSWs (three of the five) than the vortex displacement MSSWs (the other two). It is also shown that the variable forecasts among different lead times and models for each MSSW are characterized by how well the models reproduce the weakenings and deformations of the stratospheric polar vortex around the MSSWs and the intensifications of tropospheric planetary waves before them. The variability is well captured by planetary wave activity, or poleward heat flux, in the lower stratosphere. This underlines the importance of enhanced planetary waves in the troposphere and lower stratosphere for successful forecasts of the forecasting MSSWs, although the causality between the MSSWs and enhanced planetary waves will deserve further investigation.

Predictability of Northern Hemisphere Final Stratospheric Warmings and Their Surface Impacts

“…Thus, the predictive time scales of early FWs more closely match those of other dynamic stratospheric events such as midwinter SSWs (A. Y. Karpechko, 2018;Taguchi, 2018). The decreased predictability at longer lead times for wave-driven events is associated with the deterministic prediction limits of synoptic tropospheric variability (Buizza & Leutbecher, 2015;Domeisen et al, 2017) that drive these events in the stratosphere.…”

“…If we now consider only those times in which the prediction systems accurately (within ±3 days) predict the observed FW (Figure 3a), we find on average 45% of ensemble members are able to predict the observed FW at lead times of up to 20 days (note that the false alarm rates, in which a FW was predicted but did not occur, are on average about 18%-see supporting information for a description). Even out to 30-day lead times, some systems have >25% of ensemble members accurately predicting the FW, which is higher than the predictive skill at these lead times for most midwinter SSWs (Karpechko, 2018;Taguchi, 2018;A. Y. Tripathi et al, 2014).…”

“…These surface impacts often project onto the negative phase of the North Atlantic Oscillation (NAO; Butler et al, 2017;Charlton-Perez et al, 2018;Domeisen, 2019), representing an equatorward shift of the North Atlantic storm track. Recent studies have indicated that midwinter SSWs are generally only predictable at lead times of 10-15 days (Karpechko, 2018;Taguchi, 2018), though there is large variability in the predictability between events, and probabilistic prediction may be possible at weeks or even a season ahead of time (Scaife et al, 2016).…”

The Role of the Stratosphere in Subseasonal to Seasonal Prediction: 1. Predictability of the Stratosphere

“…In a common classification, there are two major types of midwinter SSW events: (1) "split" events, for which the polar vortex splits into two separate vortices, and (2) "displacement" events, for which the polar vortex is distorted and displaced off the pole (e.g., Charlton & Polvani, 2007). Taguchi (2018) provides an analysis of the predictability in the S2S hindcasts of five SSW events (December 1998, December 2001, January 2009, January 2013in the NH, and September 2002, showing that the vortex split SSWs (i.e., 2002SSWs (i.e., , 2009SSWs (i.e., , and 2013 were more difficult to forecast than the displacements (1998 and 2001). Here, we extend that analysis by considering the predictability of 11 NH midwinter SSW events in ERA-Interim during the 1996-2010 period.…”

Dependence of sudden stratospheric warming type‐transition on preceding North Atlantic Oscillation conditions

“…Since SSWs are known to be related to variability in tropospheric circulation and weather (Baldwin and Dunkerton, , ; Thompson et al ., ), predicting the occurrence of SSWs is an important issue in subseasonal to seasonal forecasting (Karpechko et al ., ; Taguchi, ; Rao et al ., ). The ability to forecast stratosphere‐troposphere coupling after warming events could be improved by understanding what mechanisms influence the change in the stratospheric polar vortex in advance.…”