Evolution of ECMWF sub-seasonal forecast skill scores

2014 ◽  
Vol 140 (683) ◽  
pp. 1889-1899 ◽  
Author(s):  
Frédéric Vitart
Keyword(s):  
Seasonal Forecast ◽  
Forecast Skill ◽  
Skill Scores

scholarly journals Seasonal Forecasts of the Twentieth Century

2020 ◽  
Vol 101 (8) ◽  
pp. E1413-E1426 ◽  
Author(s):  
Antje Weisheimer ◽  
Daniel J. Befort ◽  
Dave MacLeod ◽  
Tim Palmer ◽  
Chris O’Reilly ◽  
...  
Keyword(s):  
Twentieth Century ◽  
Large Scale ◽  
Seasonal Forecast ◽  
Forecast Skill ◽  
Seasonal Forecasts ◽  
Climate Anomalies ◽  
Global Circulation Models ◽  
Skill Scores ◽  
Physically Based ◽  
Forecast Models

Abstract Forecasts of seasonal climate anomalies using physically based global circulation models are routinely made at operational meteorological centers around the world. A crucial component of any seasonal forecast system is the set of retrospective forecasts, or hindcasts, from past years that are used to estimate skill and to calibrate the forecasts. Hindcasts are usually produced over a period of around 20–30 years. However, recent studies have demonstrated that seasonal forecast skill can undergo pronounced multidecadal variations. These results imply that relatively short hindcasts are not adequate for reliably testing seasonal forecasts and that small hindcast sample sizes can potentially lead to skill estimates that are not robust. Here we present new and unprecedented 110-year-long coupled hindcasts of the next season over the period 1901–2010. Their performance for the recent period is in good agreement with those of operational forecast models. While skill for ENSO is very high during recent decades, it is markedly reduced during the 1930s–1950s. Skill at the beginning of the twentieth century is, however, as high as for recent high-skill periods. Consistent with findings in atmosphere-only hindcasts, a midcentury drop in forecast skill is found for a range of atmospheric fields, including large-scale indices such as the NAO and the PNA patterns. As with ENSO, skill scores for these indices recover in the early twentieth century, suggesting that the midcentury drop in skill is not due to a lack of good observational data. A public dissemination platform for our hindcast data is available, and we invite the scientific community to explore them.

Assessing reasons for skilful predictions of winter windstorms over the Atlantic/European region

2020 ◽  
Author(s):  
Lisa Degenhardt ◽  
Gregor Leckebusch ◽  
Adam Scaife
Keyword(s):  
Large Scale ◽  
Surface Wind ◽  
Forecast Model ◽  
Surface Wind Speed ◽  
Seasonal Forecast ◽  
Forecast Skill ◽  
Seasonal Forecasts ◽  
Financial Industry ◽  
Near Surface ◽  
Skill Scores

<p>Severe Atlantic winter storms are affecting densely populated regions of Europe (e.g. UK, France, Germany, etc.). Consequently, different parts of the society, financial industry (e.g., insurance) and last but not least the general public are interested in skilful forecasts for the upcoming storm season (usually December to March). To allow for a best possible use of steadily improved seasonal forecasts, the understanding which factors contribute to realise forecast skill is essential and will allow for an assessment whether to expect a forecast to be skilful or not.</p><p>This study analyses the predictability of the seasonal forecast model of the UK MetOffice, the GloSea5. Windstorm events are identified and tracked following Leckebusch et al. (2008) via the exceedance of the 98<sup>th</sup> percentile of the near surface wind speed.</p><p>Seasonal predictability of windstorm frequency in comparison to observations (based e.g., on ERA5 reanalysis) are calculated and different statistical methods (skill scores) are compared.</p><p>Large scale patterns (e.g., NAO, AO, EAWR, etc.) and dynamical factors (e.g., Eady Growth Rate) are analysed and their predictability is assessed in comparison to storm frequency forecast skill. This will lead to an idea how the forecast skill of windstorms is depending on the forecast skill of forcing factors conditional to the phase of large-scale variability modes. Thus, we deduce information, which factors are most important to generate seasonal forecast skill for severe extra-tropical windstorms.</p><p>The results can be used to get a better understanding of the resulting skill for the upcoming windstorm season.</p>

scholarly journals Seasonal forecast verification and application in times of change

2016 ◽  
Author(s):  
Yoav Levi ◽  
Itzhak Carmona
Keyword(s):  
Warming Trend ◽  
Seasonal Forecast ◽  
Forecast Skill ◽  
Decision Makers ◽  
Economic Environment ◽  
Boreal Summer ◽  
Seasonal Forecasts ◽  
Climate Services ◽  
The Past ◽  
Skill Scores

Abstract. Seasonal forecast is being promoted as one of the climate services given to the public and decision makers also in the extra-tropics. However seasonal forecast is a scientific challenge. Rapid changes in climate and the socio-economic environment in the past 30 years introduce even a bigger challenge for the end-users of seasonal forecasts based on the past 30 years. Decision makers should relay on a forecast only if they fully understand the forecast skill and the forecast will not be a completely erroneous.Therefore, the percentage of forecasts for above normal condition that realized to be below normal conditions and vice versa is measured straightforwardly by the "Fiasco score". To overcome the climate and socio-economic environment changes an attempt to relate the next seasonal forecast to the previous season forecast and observed values was tested.The findings indicate that ECMWF system-4 seasonal forecast skill for June-July-August (JJA) temperatures for the marine tropics is very promising as indicated by all the skill scores, including using the previous JJA forecast as the base for the next JJA season. However for the boreal summer temperatures forecast over land, the main source of the model predictability originates from the warming trend along the hindcast period. Over the Middle East and Mongolia removing the temperature trend eliminated the high forecast skill. Evaluation of the ability of the next season forecast to predict the changes relative to the previous year's season has shown a positive skill in some areas compared to the traditional 30 years based climatology after both forecasts and observed data were de-trend.

scholarly journals Application of a Hybrid Statistical–Dynamical System to Seasonal Prediction of North American Temperature and Precipitation

2019 ◽  
Vol 147 (2) ◽  
pp. 607-625 ◽  
Author(s):  
Sarah Strazzo ◽  
Dan C. Collins ◽  
Andrew Schepen ◽  
Q. J. Wang ◽  
Emily Becker ◽  
...  
Keyword(s):  
North American ◽  
Seasonal Forecast ◽  
Forecast Skill ◽  
Precipitation Forecast ◽  
Dynamical Model ◽  
Temperature Pattern ◽  
Seasonal Forecasts ◽  
Temperature And Precipitation ◽  
Skill Scores ◽  
Calibration Models

Abstract Recent research demonstrates that dynamical models sometimes fail to represent observed teleconnection patterns associated with predictable modes of climate variability. As a result, model forecast skill may be reduced. We address this gap in skill through the application of a Bayesian postprocessing technique—the calibration, bridging, and merging (CBaM) method—which previously has been shown to improve probabilistic seasonal forecast skill over Australia. Calibration models developed from dynamical model reforecasts and observations are employed to statistically correct dynamical model forecasts. Bridging models use dynamical model forecasts of relevant climate modes (e.g., ENSO) as predictors of remote temperature and precipitation. Bridging and calibration models are first developed separately using Bayesian joint probability modeling and then merged using Bayesian model averaging to yield an optimal forecast. We apply CBaM to seasonal forecasts of North American 2-m temperature and precipitation from the North American Multimodel Ensemble (NMME) hindcast. Bridging is done using the model-predicted Niño-3.4 index. Overall, the fully merged CBaM forecasts achieve higher Brier skill scores and better reliability compared to raw NMME forecasts. Bridging enhances forecast skill for individual NMME member model forecasts of temperature, but does not result in significant improvements in precipitation forecast skill, possibly because the models of the NMME better represent the ENSO–precipitation teleconnection pattern compared to the ENSO–temperature pattern. These results demonstrate the potential utility of the CBaM method to improve seasonal forecast skill over North America.

scholarly journals Yearly evolution of Euro-Atlantic weather regimes and of their sub-seasonal predictability

2021 ◽  
Author(s):  
Nicola Cortesi ◽  
Verónica Torralba ◽  
Llorenó Lledó ◽  
Andrea Manrique-Suñén ◽  
Nube Gonzalez-Reviriego ◽  
...  
Keyword(s):  
Annual Cycle ◽  
Winter Season ◽  
Skill Score ◽  
Seasonal Forecast ◽  
Forecast Skill ◽  
Seasonal Forecasts ◽  
Sea Level Pressure ◽  
Weather Regimes ◽  
Start Dates ◽  
Weekly Frequency

AbstractIt is often assumed that weather regimes adequately characterize atmospheric circulation variability. However, regime classifications spanning many months and with a low number of regimes may not satisfy this assumption. The first aim of this study is to test such hypothesis for the Euro-Atlantic region. The second one is to extend the assessment of sub-seasonal forecast skill in predicting the frequencies of occurrence of the regimes beyond the winter season. Two regime classifications of four regimes each were obtained from sea level pressure anomalies clustered from October to March and from April to September respectively. Their spatial patterns were compared with those representing the annual cycle. Results highlight that the two regime classifications are able to reproduce most part of the patterns of the annual cycle, except during the transition weeks between the two periods, when patterns of the annual cycle resembling Atlantic Low regime are not also observed in any of the two classifications. Forecast skill of Atlantic Low was found to be similar to that of NAO+, the regime replacing Atlantic Low in the two classifications. Thus, although clustering yearly circulation data in two periods of 6 months each introduces a few deviations from the annual cycle of the regime patterns, it does not negatively affect sub-seasonal forecast skill. Beyond the winter season and the first ten forecast days, sub-seasonal forecasts of ECMWF are still able to achieve weekly frequency correlations of r = 0.5 for some regimes and start dates, including summer ones. ECMWF forecasts beat climatological forecasts in case of long-lasting regime events, and when measured by the fair continuous ranked probability skill score, but not when measured by the Brier skill score. Thus, more efforts have to be done yet in order to achieve minimum skill necessary to develop forecast products based on weather regimes outside winter season.

Interannual Variations and Prediction of Spring Precipitation over China

2018 ◽  
Vol 31 (2) ◽  
pp. 655-670 ◽  
Author(s):  
YuJia You ◽  
Xiaojing Jia
Keyword(s):  
Regression Model ◽  
Yangtze River ◽  
Climate Models ◽  
Seasonal Forecast ◽  
Forecast Skill ◽  
Interannual Variations ◽  
The Yangtze River ◽  
Spring Precipitation ◽  
Coupled Climate Models ◽  
Time Variations

The interannual variations and the prediction of the leading two empirical orthogonal function (EOF) modes of spring (April–May) precipitation over China for the period from 1951 to 2014 are investigated using both observational data and the seasonal forecast made by six coupled climate models. The leading EOF mode of spring precipitation over China (EOF1-prec) features a monosign pattern, with the maximum loading located over southern China. The ENSO-related tropical Pacific SST anomalies in the previous winter can serve as a precursor for EOF1-prec. The second EOF mode of spring precipitation (EOF2-prec) over China is characterized by a dipole structure, with one pole near the Yangtze River and the other one with opposite sign over the Pearl River delta. A North Atlantic sea surface temperature (SST) anomaly dipole in the preceding March is found contribute to the prec-EOF2 and can serve as its predictor. A physics-based empirical (P-E) model is then formulated using the two precursors revealed by the observational analysis to forecast the variations of EOF1-prec and EOF2-prec. Compared to coupled climate models, which have little skill in forecasting the time variations of the two EOF modes, this P-E model can significantly improve the forecast skill of their time variations. A linear regression model is further established using the time series forecast by the P-E model to forecast the spring precipitation over China. Results suggest that the seasonal forecast skill of the spring precipitation over southeastern China, especially over the Yangtze River area, can be significantly improved by the regression model.

scholarly journals Enhanced seasonal forecast skill following stratospheric sudden warmings

2013 ◽  
Vol 6 (2) ◽  
pp. 98-102 ◽  
Author(s):  
M. Sigmond ◽  
J. F. Scinocca ◽  
V. V. Kharin ◽  
T. G. Shepherd
Keyword(s):  
Seasonal Forecast ◽  
Forecast Skill ◽  
Stratospheric Sudden Warmings

On the Level and Origin of Seasonal Forecast Skill in Northern Europe

1998 ◽  
Vol 55 (1) ◽  
pp. 103-127 ◽  
Author(s):  
Åke Johansson ◽  
Anthony Barnston ◽  
Suranjana Saha ◽  
Huug van den Dool
Keyword(s):  
Seasonal Forecast ◽  
Forecast Skill ◽  
Northern Europe

Diagnosing Sources of U.S. Seasonal Forecast Skill

2006 ◽  
Vol 19 (13) ◽  
pp. 3279-3293 ◽  
Author(s):  
X. Quan ◽  
M. Hoerling ◽  
J. Whitaker ◽  
G. Bates ◽  
T. Xu
Keyword(s):  
Air Temperature ◽  
Southern Oscillation ◽  
Surface Air Temperature ◽  
Seasonal Forecast ◽  
Forecast Skill ◽  
Diagnostic Methods ◽  
Sea Surface ◽  
Empirical Methods ◽  
Spatial Coverage ◽  
Sst Anomalies

Abstract In this study the authors diagnose the sources for the contiguous U.S. seasonal forecast skill that are related to sea surface temperature (SST) variations using a combination of dynamical and empirical methods. The dynamical methods include ensemble simulations with four atmospheric general circulation models (AGCMs) forced by observed monthly global SSTs from 1950 to 1999, and ensemble AGCM experiments forced by idealized SST anomalies. The empirical methods involve a suite of reductions of the AGCM simulations. These include uni- and multivariate regression models that encapsulate the simultaneous and one-season lag linear connections between seasonal mean tropical SST anomalies and U.S. precipitation and surface air temperature. Nearly all of the AGCM skill in U.S. precipitation and surface air temperature, arising from global SST influences, can be explained by a single degree of freedom in the tropical SST field—that associated with the linear atmospheric signal of El Niño–Southern Oscillation (ENSO). The results support previous findings regarding the preeminence of ENSO as a U.S. skill source. The diagnostic methods used here exposed another skill source that appeared to be of non-ENSO origins. In late autumn, when the AGCM simulation skill of U.S. temperatures peaked in absolute value and in spatial coverage, the majority of that originated from SST variability in the subtropical west Pacific Ocean and the South China Sea. Hindcast experiments were performed for 1950–99 that revealed most of the simulation skill of the U.S. seasonal climate to be recoverable at one-season lag. The skill attributable to the AGCMs was shown to achieve parity with that attributable to empirical models derived purely from observational data. The diagnostics promote the interpretation that only limited advances in U.S. seasonal prediction skill should be expected from methods seeking to capitalize on sea surface predictors alone, and that advances that may occur in future decades could be readily masked by inherent multidecadal fluctuations in skill of coupled ocean–atmosphere systems.

scholarly journals Origins and Levels of Seasonal Forecast Skill for Sea Ice in Hudson Bay Using Canonical Correlation Analysis

2011 ◽  
Vol 24 (5) ◽  
pp. 1378-1395 ◽  
Author(s):  
Adrienne Tivy ◽  
Stephen E. L. Howell ◽  
Bea Alt ◽  
John J. Yackel ◽  
Thomas Carrieres
Keyword(s):  
Correlation Analysis ◽  
Sea Ice ◽  
Canonical Correlation Analysis ◽  
North Atlantic ◽  
Canonical Correlation ◽  
Seasonal Forecast ◽  
Forecast Skill ◽  
Sea Surface ◽  
Hudson Bay ◽  
Ice Concentration

Abstract Canonical correlation analysis (CCA) is used to estimate the levels and sources of seasonal forecast skill for July ice concentration in Hudson Bay over the 1971–2005 period. July is an important transition month in the seasonal cycle of sea ice in Hudson Bay because it is the month when the sea ice clears enough to allow the first passage of ships to the Port of Churchill. Sea surface temperature (quasi global, North Atlantic, and North Pacific), Northern Hemisphere 500-mb geopotential height (z500), sea level pressure (SLP), and regional surface air temperature (SAT) are tested as predictors at 3-, 6-, and 9-month lead times. The model with the highest skill has three predictors—fall North Atlantic SST, fall z500, and fall SAT—and significant tercile forecast skill covering 61% of the Hudson Bay region. The highest skill for a single-predictor model is from fall North Atlantic SST (6-month lead). Fall SST explains 69% of the variance in July ice concentration in Hudson Bay and a possible atmospheric link that accounts for the lagged relationship is presented. CCA diagnostics suggest that changes in the subpolar North Atlantic gyre and the Atlantic multidecadal oscillation (AMO), reflected in sea surface temperature, precedes a deepening/weakening of the winter upper-air ridge northwest of Hudson Bay. Changes in the height of the ridge are reflected in the strength of the winter northwesterly winds over Hudson Bay that have a direct impact on the winter ice thickness distribution; anomalies in winter ice severity are later reflected in the pattern and timing of spring breakup. July ice concentration in Hudson Bay has declined by approximately 20% per decade between 1979 and 2007, and the hypothesized link to the AMO may help explain this significant loss of ice.

Sign in / Sign up

Export Citation Format

Share Document