4-9 September 2022, Bonn, Germany
Enter zoom meeting

UP3.7

Sub-seasonal to seasonal predictability: Processes, methods, and impacts

Prediction and predictability on timescales of several weeks to months is crucial for the advancement of our understanding and modeling of processes on these timescales. These processes include coupling processes in the global climate system, their representation and prediction in model systems, as well as the impacts associated with extreme events that exhibit probabilistic predictability on these timescales. This session invites contributions that span all aspects of prediction and predictability in the lead time range between 2 weeks and seasonal timescales. We encourage submissions on physical processes, including (but not limited to) the Madden Julian Oscillation (MJO), the monsoons, and El Nino Southern Oscillation (ENSO) and their remote effects, coupling between different parts of the globe, the vertical coupling in the atmosphere, as well as coupling between the atmosphere and the underlying surface in terms of land, ocean and the cryosphere. We further invite contributions on ensemble prediction and analysis methods as well as impact-based methods for socio-economic impacts related to processes and predictability on sub-seasonal to seasonal timescales.

Convener: Daniela Domeisen | Co-conveners: Johanna Baehr, Dominik Büeler, Maria Pyrina, Frederic Vitart, Christopher White, Priyanka Yadav
Orals
| Thu, 08 Sep, 11:00–13:00 (CEST)|Room HS 1
Posters
| Wed, 07 Sep, 16:00–17:15 (CEST) | Display Wed, 07 Sep, 08:00–18:00|b-IT poster area

Thu, 8 Sep, 11:00–13:00

Chairpersons: Priyanka Yadav, Dominik Büeler, Daniela Domeisen

11:00–11:15
|
EMS2022-156
|
solicited
|
Onsite presentation
Marisol Osman et al.

The prediction skill of sub-seasonal forecast models is evaluated for  seven year-round weather regimes in the Atlantic-European region, with a focus on regime onsets and transitions which modulate surface weather in a way that is particularly relevant for the European energy system. Re-forecasts using models from three prediction centers (the European Centre for Medium-Range Weather Forecasts, the National Center for Environmental Prediction and the UK Met Office) for the 2000-2015 period are considered and compared against weather regimes obtained from ERA Interim reanalysis over the same period. We first evaluate their ability to reproduce weather regime life-cycle characteristics, such as their frequency, length, number and transitions. Then, we focus on the assessment of skill, placing emphasis on the differences in the performance for each weather regime depending on the time of the year. Finally, we consider the year-to-year evolution of skill and the role of interannual variability of the atmosphere in this skill.

Results show that the largest biases in frequency are obtained for Scandinavian Blocking in summer due to an underestimation of the number of life cycles for this regime. The ECMWF model shows the highest skill for most of the weather regimes and seasons, followed closely by the NCEP model. The average regime skill horizon is 3 days longer for ECMWF and NCEP models than for the UKMO model, mainly due to the differences in skill in winter. Greenland Blocking tends to have the longest year-round skill horizon for the three models driven by their performance in winter, which is skillful into week 3 of the forecast period. On the other hand, the skill is lowest for the European Blocking regime for the three models, followed by Scandinavian Blocking. These results demonstrate that weather regime forecasts have the potential to identify periods that may exhibit enhanced forecast skill at sub-seasonal timescales, while at the same time skill depends upon the specific regime.

How to cite: Osman, M., Grams, C. M., and Beerli, R.: Sub-seasonal prediction of the year-round Atlantic-European weather regimes, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-156, https://doi.org/10.5194/ems2022-156, 2022.

11:15–11:30
|
EMS2022-226
|
CC
|
Onsite presentation
Jan Wandel et al.

Variability of the large-scale atmospheric circulation in the midlatitudes is characterised by transient extratropical Rossby waves and stationary patterns, such as blocking anticyclones. On sub-seasonal time scales, the extratropical variability can be depicted by weather regimes which are quasi-stationary, persistent, and recurrent flow patterns. Accurate forecasts of the weather regimes are particularly relevant due to their longevity on sub-seasonal time scales and potential socio-economic impact. Research in recent years emphasized the modulation of blocked weather regimes by processes on synoptic scales, in particular by diabatic processes in extratropical cyclones. These diabatic processes are related to latent heat release in the so-called warm conveyor belt (WCB). The WCB is an ascending air stream in the warm sector of an extratropical cyclone and its “diabatic outflow” modulates the upper-level jet, resulting in the amplification of an upper-level ridge and eventually a block. This study systematically investigates the representation of WCBs around the onset of blocked weather regimes over the Atlantic-European region in the sub-seasonal reforecasts of the European Centre for Medium-Range Weather Forecasts (ECMWF) Integrated Forecasting System (IFS). In general, there is a strong link between the flow amplification around blocking onsets and high WCB outflow activity in the region of the incipient block. The analysis of the 11 ensemble members in the extended winter season from 1997 – 2017 shows that the ECMWF’s IFS reforecasts underestimate both the wave amplification and WCB frequencies for regime onsets in forecast week 2 (forecast day 8 – 14). Due to the persistence of the weather regimes, onsets on these time scales typically lead to active life cycles in week 3 and sometimes even week 4. The analysis of the temporal evolution prior to the onset reveals that the reforecasts establish large-scale flow anomalies via WCB outflow differently than observed with a more amplified large-scale flow and high WCB activity upstream of the incipient block. Our findings show that forecast biases in WCB activity might be partly responsible for the relatively poor skill of blocking forecasts in week 2 and 3 and highlight the potential for further improvement of sub-seasonal prediction by improving the representation of WCBs in NWP models.

 

How to cite: Wandel, J., Quinting, J. F., Büeler, D., Knippertz, P., and Grams, C. M.: The role of warm conveyor belts for the sub-seasonal prediction of blocked weather regimes, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-226, https://doi.org/10.5194/ems2022-226, 2022.

11:30–11:45
|
EMS2022-429
|
Onsite presentation
Muhammad Adnan Abid et al.

The North Atlantic European (NAE) regional circulation anomalies variability and predictability during the boreal winter season is modulated through both stratospheric and tropospheric pathways on sub-seasonal to seasonal timescales. El Niño-Southern Oscillation (ENSO) through the atmospheric teleconnections is one of the dominating forcing that modulates the global climate variability and Predictability during the boreal winter season. However, recent studies indicate the intra-seasonality in the ENSO teleconnections, where through the inter-basin interactions the Indian Ocean (IO) bridges the ENSO response to the NAE region in the early winter, while the direct ENSO response dominates in latter half of the boreal winter. Therefore, in the current study we analyzed the predictability of the NAE circulation anomalies and the tropical Indian Ocean precipitation anomalies during the early winter season using the ECMWF System-5 seasonal (SEAS5) dataset. We noted the boreal Autumn Indian Ocean Dipole (IOD) conditions are the pre-courser for the early winter precipitation anomalies in the Tropical Western-Central Indian Ocean (TWCIO) region, which dominates the ENSO response to the IO precipitation anomalies during the earlier half of the winter season. These TWCIO precipitation anomalies are well predicted by the ECMWF-SEAS5 prediction system during early winter. Furthermore, we noted the positive TWCIO heating anomalies tend to favor the positive North Atlantic Oscillation (NAO) condition in the North Atlantic region. This leads to the above normal Surface Air temperature (SAT) conditions, indicating to the mild early winter conditions over the European continent. The ECMWF-SEAS5 system shows a significant prediction skill of the SAT anomalies over the NAE region.

How to cite: Abid, M. A., Kucharski, F., and Molteni, F.: Predictability of the North Atlantic European region and the role of Indian Ocean during early winter, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-429, https://doi.org/10.5194/ems2022-429, 2022.

11:45–12:00
|
EMS2022-192
|
Online presentation
Chaim Garfinkel and Dvir Chwat
The predictability of sudden stratospheric  warming (SSW) events are considered in 10 subseasonal to seasonal (S2S) forecast models for 16 SSWs over the period 1998 - 2021.  The four factors that most succinctly distinguish those SSWs with above average predictability are an active Madden-Julian Oscillation  with enhanced convection in the West Pacific, the Quasi-Biennial Oscillation phase with easterlies in the lower stratosphere,  a strong pulse of wave activity   before the event, and the morphology (displacement more predictable). However none of these effects are statistically significant at the 95\% level using a two-tailed t-test due to  the relatively small sample size. Strong events are more predictable if one focuses on the question of whether the models predict a SSW, though not if one focuses on the absolute error of the anomalous stratospheric easterlies. Other factors, such as El Nino and  stratospheric preconditioning are comparatively less important. 

How to cite: Garfinkel, C. and Chwat, D.: Which Sudden Stratospheric  Warming Events are Most Predictable?, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-192, https://doi.org/10.5194/ems2022-192, 2022.

12:00–12:15
|
EMS2022-69
|
Online presentation
Clementine Dalelane

The variability of the sea level pressure in the North Atlantic sector is the most important driver of weather and climate in Europe. The main mode of this variability, the North Atlantic Oscillation (NAO), explains up to 50% of the total variance. Other modes, known as the Scandinavian index, East Atlantic and East Atlantic/West Russian pattern, complement the variability of the sea level pressure, thereby influencing the European climate.

Current seasonal forecasts of European winter climate, though highly desirable for society and economy, are as yet not fully reliable. There exist a number of autumn predictors, such as sea surface and stratospheric temperature, Eurasian snow depth, and Arctic sea ice extension, that impact on the upcoming pressure regimes in a predictable way. The present dynamical seasonal forecast systems respond still too weakly to these known seasonal predictors. But the relationship is reproduced quite well by means of statistics.

In combination, statistical and dynamical forecasts have the potential to improve forecasts of the North Atlantic pressure conditions and thereby affected variables like temperature and precipitation in Europe considerably. A seasonal prediction system with enhanced winter NAO skill due to ensemble subsampling w.r.t. a statistical estimate of the NAO index entails an improved prediction of the surface climate variables as well. Here, we show that a refined subselection procedure that accounts both for the NAO index and for the three additional modes of sea level pressure variability, is able to further increase the prediction skill of the operational seasonal forecast model of the German Meteorological Service GCFS of wintertime mean sea level pressure, near-surface temperature and precipitation across Europe.     

 

How to cite: Dalelane, C.: Seasonal Forecasts of Winter Temperature Improved by Higher-Order Modes of Mean Sea Level Pressure Variability in the North Atlantic Sector, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-69, https://doi.org/10.5194/ems2022-69, 2022.

12:15–12:30
|
EMS2022-110
|
Online presentation
Niclas Rieger et al.

The last few years have seen an ever growing interest in weather predictions on sub-seasonal time scales ranging from 2 weeks to about 2 months. By forecasting aggregated weather statistics, such as weekly precipitation, it has indeed become possible to overcome the theoretical predictability limit of 2 weeks, bringing life to time scales which historically have been known as the “predictability desert”. The growing success at these time scales is largely due to the identification of weather and climate processes providing sub-seasonal predictability, such as the Madden-Julian Oscillation (MJO) and anomaly patterns of global sea surface temperature (SST), sea surface salinity (SSS), soil moisture and snow cover. Although much has been gained by these studies, a comprehensive analysis of potential predictors and their relative relevance to forecast sub-seasonal rainfall is still missing.

At the same time, data-driven machine learning (ML) models have proved to be excellent candidates to tackle two common challenges in weather forecasting: (i) resolving the non-linear relationships inherent to the chaotic climate system and (ii) handling the steadily growing amounts of Earth observational data. Not surprisingly, a variety of studies have already displayed the potential of ML models to improve the state-of-the-art dynamical weather prediction models currently in use for sub-seasonal predictions, in particular for temperatures, precipitation and the MJO. It seems therefore inevitable that the future of sub-seasonal prediction lies in the combination of both the dynamical, process-based and the statistical, data-driven approach. 

In the advent of this new age of combined Neural Earth System Modeling, we want to provide insight and guidance for future studies (i) to what extent large-scale teleconnections on the sub-seasonal scale can be resolved by purely data-driven models and (ii) what the relative contributions of the individual large-scale predictors are to make a skillful forecast. To this end, we build neural networks to predict sub-seasonal precipitation based on a variety of large-scale predictors derived from oceanic, atmospheric and terrestrial sources. As a second step, we apply layer-wise relevance propagation to examine the relative importance of different climate modes and processes in skillful forecasts.

Preliminary results show that the skill of our data-driven ML approach is comparable to state-of-the-art dynamical models suggesting that current operational models are able to correctly model large-scale teleconnections within the climate system. The ML model achieves highest skills over the tropical Pacific and the western part of North America. By investigating the relative importance of those large-scale predictors for skillful predictions, we find that the MJO and slow-varying processes associated with SST and SSS anomalies like the El Niño-Southern Oscillation, the Pacific decadal oscillation and the Atlantic meridional mode all play an important role for individual regions.

How to cite: Rieger, N., Olmedo, E., Corral, Á., Magnusson, L., Ferranti, L., and Turiel, A.: Identifying relevant large-scale predictors for sub-seasonal precipitation forecast using explainable neural networks, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-110, https://doi.org/10.5194/ems2022-110, 2022.

12:30–12:45
|
EMS2022-497
|
Onsite presentation
Meriem Krouma et al.

The Madden-Julian Oscillation (MJO) is one of the main sources of sub-seasonal atmospheric predictability in the Tropical region. The MJO affects precipitation over highly populated areas, especially around Southern India. Therefore, predicting its phase and intensity is a scientific challenge of high societal impact.

Indices of the MJO can be derived from the first principal components of wind speed and outgoing longwave radiation (OLR) in the Tropics (RMM1 and RMM2 indices). The amplitude and phase of the MJO are derived from those indices. Our challenge is to forecast these two indices on a sub-seasonal timescale. This study aims to provide an ensemble forecast of MJO indices from analogs of the atmospheric circulation, computed from the geopotential at 500 hPa (Z500) by using a stochastic weather generator (SWG). The SWG is based on the random sampling of circulation analogs, which is a simple form of machine learning simulation.

We generate an ensemble of 100 members for the MJO amplitude and the RMMs for sub-seasonal lead times (from 2 to 4 weeks). Then we evaluate the skill of the ensemble forecast and the ensemble mean using respectively probabilistic (CRPSS) and deterministic skill scores (correlation and RMSE). We found that a reasonable forecast could be achieved for 40-day lead times for the different seasons. We compare our SWG approach with other forecasts of the MJO mainly with the ECMWF forecast and machine learning forecast. We found that the SWG has reliable forecast skills compared to other forecasts in particular for lead times up to 20 days.

How to cite: Krouma, M., Yiou, P., and Silini, R.: Ensemble forecast of an index of the Madden Julian Oscillation using a stochastic weather generator based on analogs of  Z500, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-497, https://doi.org/10.5194/ems2022-497, 2022.

12:45–13:00
|
EMS2022-434
|
Onsite presentation
Maria Madsen et al.

Commonly studied atmospheric phenomena are the persistent, quasi-stationary high-pressure systems known as atmospheric blocking. During the boreal winter, these blocking anticyclones often lead to downstream cold temperature extremes and have been associated with processes varying on longer sub-seasonal timescales such as the Madden Julian oscillation (MJO) and the stratospheric polar vortex. The Atlantic-European region has recently been characterized by four types of blocked regimes, each with differing downstream sensible weather impacts and forecast skill. Key to better predictability of these Atlantic-European blocked weather regimes is identifying remote influences on sub-seasonal timescales.

Recently, linear inverse models (LIMs) have been successfully used to quantify slowly varying precursors to large-scale circulations patterns in the Northern Hemisphere. A LIM is an empirical model using lag covariance statistics to approximate dynamical properties and the evolution of specific atmospheric variables contained in a prescribed atmospheric state vector. The evolution of the state vector is approximated by separating slowly evolving linearized dynamics, as well as a linear approximation to nonlinear dynamics, from presumably unpredictable white-noise forcing. In this work, a LIM is employed to identify optimal precursors on sub-seasonal time scales for the life-cycle of Atlantic-European blocking events during the extended boreal winter (November-March). Included in the LIM are tropical heating anomalies related to the MJO or the El Niño-Southern Oscillation (ENSO), extratropical influences from Rossby wave activity, and the stratospheric polar vortex. Optimal extratropical and tropical precursors for the varying types of blocking events, as well as the LIM’s skill in forecasting European-Atlantic blocking, are presented.  

How to cite: Madsen, M., Wirth, V., Riemer, M., and Grams, C.: Characterizing Optimal Atlantic-European Blocking Precursors Using a Linear Inverse Model , EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-434, https://doi.org/10.5194/ems2022-434, 2022.

Posters

P47
|
EMS2022-311
|
Onsite presentation
Thang M Luong et al.

Despite being one of the driest places in the world, the Kingdom of Saudi Arabia (KSA) occasionally experiences extreme precipitation events associated with organized convections that might lead to flooding. Rainfall forecasts at lead times on the sub-seasonal to seasonal (S2S) timescale can potentially assist disaster risk mitigation, and water resource management. Here, model skills of predicting precipitation at sub-seasonal scale (from 2 to 4 weeks ahead) are benchmarked over the Arabian Peninsula (AP). We utilized the Weather Research and Forecasting Model (WRF) at convective-permitting resolution (4 km) to dynamically downscale ensemble of 11 members of the European Centre of Medium-range Weather Forecasts (ECMWF) S2S reforecast product over a 20-year period (1998-2018). Representation of precipitation is assessed with a regional reanalysis over the AP and in-situ rain gauge measurements.

Precipitation mostly occurs over the AP during cooler months (November to April). Mesoscale convective systems (MCSs) are important factors in producing rainfall over the region during this period when extratropical systems are dominant. The majority of rainfall events in November to February are associated with extratropical forcing, while March and April rainfalls are associated with tropical-extratropical interactions.

Our results indicate that the WRF convective-permitting model adequately describes the precipitation patterns over the AP up to 4-week forecast-range and statistically improves the forecast skill with regard to its driving ECMWF fields over the studied 20-year period. Large-scale circulation signatures are reproduced in the model better in the spring wide-spread rainfall events for both WRF and ECMWF. WRF adds more values in simulating winter mesoscale convective systems.

How to cite: Luong, T. M., Dasari, H. P., Chang, H.-I., Risanto, C. B., Attada, R., Castro, C. L., and Hoteit, I.: Predictability of Sub-seasonal Rainfall Associated with Large-scale Circulations, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-311, https://doi.org/10.5194/ems2022-311, 2022.

P48
|
EMS2022-362
|
Onsite presentation
Taehyoun Shim et al.

This study introduces the prediction performance of the Korean Integrated Model (KIM) associated with medium-long range or sub-seasonal timescales. KIM has been developed as the Korea Meteorological Administration’s operational numerical weather prediction (NWP) system by the Korea Institute of Atmospheric Prediction Systems (KIAPS). KIM is a newly introduced global atmospheric model system, consisting of a spectral-element non-hydrostatic dynamical core on a cubed sphere grid and an advanced physics parameterization package (Kim et al., 2021; Hong et al., 2018). We’ll conduct hindcast sets and assess the general characteristics and systematic biases of KIM focused on a sub-seasonal timescale.

Through analyzing KIM’s hindcast experiments, we try to examine the prediction performance and predictability in the lead time range between 2 weeks and sub-seasonal timescales. This timescale is one of the important issues in terms of a seamless forecast that links NWP and climate prediction. A seamless forecast means bridging discrete short-term weather forecasts valid at a specific time and time-averaged forecast at longer periods. Sub-seasonal predictions span this time range and this transition period determines forecasting skills.

We plan to perform KIM’s ensemble reforecast simulation for the boreal winter and summer cases for the period of 2001 – 2020 (20 years). The hindcast results are compared with reanalysis data (ERA5). In order to evaluate KIM’s sub-seasonal performance, several skill scores are calculated, which are Root Mean Square Error (RMSE), Anomaly Correlation Coefficient (ACC), Ratio of Predictable Components (RPC), and Root Mean Square Skill Score (RMSSS). Finally, it is expected that these multi-year simulations will contribute to improving sub-seasonal and seasonal predictabilities.

How to cite: Shim, T., Kim, S.-W., Kim, S.-W., and Kim, J.: Characteristics of sub-seasonal prediction performance revealed on hindcast of the Korean Integrated Model, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-362, https://doi.org/10.5194/ems2022-362, 2022.

P49
|
EMS2022-389
|
Onsite presentation
Stanislava Kliegrova et al.

Long range forecasts provide information about expected future atmospheric and oceanic conditions averaged over periods of one to three months and are attractive for many sectors. They have made considerable progress in recent years, but seasonal predictability remains a problem in many regions (for example in Europe). Demand is also for long range forecasts, which would predict shorter periods than months (for example 3 decades in each month).

This study considers four seasonal forecasting systems available in the Copernicus Climate Change Service (C3S) archive which provide near-surface air temperature and precipitation data at 1°by 1°spatial resolution: European Centre for Medium-range Weather Forecast System SEAS5 (ECMWF), Météo – France System 8 (MF), Deutscher Wetterdienst GCFS 2.1 (DWD) and Centro Euro-Mediterraneo sui Cambiamenti Climatici SPSv3.5 (CMCC). It quantifies their value in predicting temperature and precipitation at monthly and shorter (3 decades in each month) temporal resolution over Europe. There are two starting dates, May 1st, and November 1st, and forecasts at lead times up to 3 months for each year in the period 1993–2016 (the longest period of hindcasts common to all systems).

We focus on 2 domains: larger (Europe, latitude 42-55°N, longitude 2-30°E) and smaller (the Czech Republic, latitude 47-52°N, longitude 11-20°E). E-OBS daily gridded observational datasets for precipitation and temperature at 0.25°spatial resolution are used as a reference.

Several statistical measures such as mean bias and root mean square error are presented for temperature and precipitation at a monthly and shorter (3 decades in each month) temporal resolution over Europe and for the Czech Republic.

How to cite: Kliegrova, S., Belda, M., Benacek, P., Metelka, L., and Stepanek, P.: Temperature and precipitation long range forecasts over Europe at a monthly and shorter temporal resolution, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-389, https://doi.org/10.5194/ems2022-389, 2022.

P50
|
EMS2022-355
|
CC
|
Onsite presentation
Dominik Büeler et al.

Extratropical cyclones strongly interact with the midlatitude waveguide and can thus actively influence onset, maintenance, and decay of large-scale weather regimes on sub-seasonal timescales (10 – 60 days). For instance, individual cyclones over the Pacific-North American region have shown to be able to trigger positive or negative phases of the North Atlantic Oscillation, depending on the type of Rossby wave breaking they are associated with. Likewise, strongly diabatically driven cyclones have shown to be relevant for ridge amplification and thus blocking onset and maintenance. Biases in cyclone activity and characteristics in sub-seasonal numerical weather prediction models might therefore hinder exploiting the potential large-scale predictability on these timescales. We thus, for the first time, identify and track extratropical cyclones in 21 winters (2000 – 2020) of sub-seasonal ensemble hindcasts from the European Centre for Medium-Range Weather Forecasts. This quasi-Lagrangian, object-oriented approach allows us to validate various cyclone life cycle characteristics such as the deepening rate, location of genesis, maximum intensity, and decay, propagation direction and speed, and size, age and lifetime. Overall, the hindcasts reproduce the climatology of cyclone activity and characteristics remarkably well up to 6 weeks lead time, both over the North Atlantic and the North Pacific. However, the hindcasts tend to underestimate the frequency of the strongly intensifying subset of North Atlantic cyclones. This underestimation is likely linked to an underestimation of cyclogenesis frequency along the southeastern U.S. coast. As strongly intensifying cyclones are particularly relevant for the reorganization of the large-scale flow downstream, this bias might influence sub-seasonal forecast skill and deserves further attention.

How to cite: Büeler, D., Sprenger, M., and Wernli, H.: Validation of extratropical cyclone characteristics in sub-seasonal ECMWF forecasts, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-355, https://doi.org/10.5194/ems2022-355, 2022.

P51
|
EMS2022-215
|
Onsite presentation
Rachel Wai-Ying Wu et al.

Extreme stratospheric events such as sudden stratospheric warmings (SSWs) and strong vortex events can have downward impacts on surface weather that can last for several weeks to months. In subseasonal-to-seasonal (S2S) prediction systems, the predictability of these events strongly differs within events of the same type, and also between event types. The reason for the differences in predictability between events is however not resolved. To investigate this question using a larger sample size, we extend the definition of strong vortex and SSW events to wind acceleration and deceleration events due to their similar dynamics, respectively. Specifically, we use the zonal mean zonal wind at 60°N, 10hPa from ERA-interim reanalysis for the winters of 1998/99 to 2017/18 to identify wind acceleration and deceleration events, which are defined as a wind change over a 10-day window. We then assess the predictability of the identified events using the hindcasts from the ECMWF S2S prediction system. Overall,  wind acceleration events are found to be more predictable than deceleration events. However, when expressing the predictability of deceleration and acceleration events as a function of event magnitude, they qualitatively exhibit the same predictability behaviour; that is, events of stronger magnitude are less predictable. We explain the observed predictability dependence from two perspectives: 1) In a statistical sense, strong magnitude events lie within the tails of the climatological distribution and thus are penalised more heavily than weak magnitude events, and 2) from a dynamical perspective, extreme stratospheric events are associated with strong anomalies in precursors such as wave activity and vortex background state, which are themselves often associated with large ensemble spread and large uncertainties. The model shows a poor predictability of extreme wave activity fluxes in the lower stratosphere. In particular, the split SSW events in 2009 and 2018, which are reported to be associated with anomalously strong wave-2 activity flux, are found to exhibit large prediction errors even at short lead times, suggesting that nonlinear wave dynamics might play an important role.  We suggest that a better representation of extremely strong wave activity in the prediction system may enhance the predictability of stratospheric extreme events, and by extension their impacts on surface weather and climate.

How to cite: Wu, R. W.-Y., Wu, Z., and Domeisen, D. I. V.: Understanding the Differences in the Sub-seasonal Predictability of Extreme Stratospheric Events: The extreme wave activity flux events in 2009 and 2018, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-215, https://doi.org/10.5194/ems2022-215, 2022.

P52
|
EMS2022-657
|
Onsite presentation
Steffen Tietsche et al.

Multi-year trends in sub-seasonal reforecasts that are inconsistent with observed trends constitute a time- or state-dependent bias in either the physical forecast model or the forecast initialization data and methods, regardless of whether the observed trends are due to climate change or slow internal variability. These reforecast trend errors degrade the skill diagnosed from the reforecasts and point to deficiencies of the sub-seasonal forecasting system in representing a changing mean state. However, detection and quantification of these trend errors is non-trivial because of often high levels of sub-seasonal to interannual variability in combination with weak trends. Here, we propose methods to assess the robustness and importance of trends in observations, and detect when trends in sub-seasonal reforecasts are inconsistent with the observed trends. As a concrete example, we pick surface air temperature (SAT) trends in Northern Hemisphere winter, and assess consistency of trends in ECMWF 47R1 reforecasts at different lead times for the 20-year period 2000-2019 with the ERA5 reanalysis trend. We find that some regions - even for this relatively short period - exhibit positive SAT trends in ERA5 that are significantly different from zero, robust against small changes in the period and relatively large when compared to interannual variability. Among these regions are the Tropical Warm Pool and the Eurasian Arctic. In the latter region, the reforecasts clearly underrepresent the observed warming trend of about 2 K per decade at longer lead times: for lead times beyond three weeks, the reforecast trend is only about 1 K per decade, with a 95% confidence interval from 0.5 to 1.5 K per decade. We discuss potential reasons for this specific trend error, with the aim to provide guidance for future improvements in the physical forecast model and data assimilation methods.

How to cite: Tietsche, S., Vitart, F., Mayer, M., and Balmaseda, M.: Multi-year trend errors in sub-seasonal reforecasts, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-657, https://doi.org/10.5194/ems2022-657, 2022.

Supporters & sponsors