Detailseite
Projekt Druckansicht

Upper-level extratropical disturbances over the low-latitude North Atlantic: Generation mechanisms and predictability

Fachliche Zuordnung Physik und Chemie der Atmosphäre
Förderung Förderung von 2007 bis 2012
Projektkennung Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 53244370
 
Erstellungsjahr 2012

Zusammenfassung der Projektergebnisse

Weather disturbances in the upper troposphere (around 7 km above sea level) frequently penetrate from the region of main storm activity over the extratropical North Atlantic into low latitudes, where they can be involved in significant weather events such as heavy precipitation over the western and northern parts of Africa, or Saharan dust outbreaks. This project is the first that systematically explored these disturbances using the huge dynamical information content of long-term data from the European Centre for Medium-Range Weather Forecasts operational ensemble prediction system (EPS) and from the recently established multi-model THORPEX Interactive Grand Global Ensemble (TIGGE). Ensemble prediction is a relatively new development in weather forecasting and aims at estimating uncertainties through multiple model runs from slightly different initial conditions or even by using different atmospheric models (as in the case of TIGGE). This study investigated the dynamics, forecast quality and predictability of low-latitude upper disturbances exploring a wide range of verification and analysis tools based on potential vorticity (PV), which is a widely used dynamical tracer for airmasses penetrating equator- and downward into the troposphere. Both statistical approaches and case studies of high-impact weather events were pursued. The main conclusions from this work are: (i) The ECMWF ensemble prediction system shows a generally underdispersive behaviour in the region of the upper disturbances. This means that the forecast errors are on average larger than the spread of the ensemble for these particular situations. (ii) Forecast errors are reduced by about 50% if each TIGGE centre’s own analysis is used instead of the ECMWF analysis, particularly in regions of large PV gradients. This result calls for more thorough reflections in future work on what to use as the ‘truth’ in forecast evaluation. (iii) The breaking of large-scale upper-level waves in the atmosphere that is often responsible for the generation of the disturbances studied here is often too weak in the forecasts, leading to a northeastward shift. (iv) Ensemble-mean multi-model forecasts of a fourday accumulated precipitation event appear accurate enough for a successful severeweather warning in contrast to some single-model ensemble predictions. (v) Condensational heating in clouds upstream early in the forecasts appear to be more important than high-pressure systems downstream for strengthening the upper disturbances in the medium range (3–7 days forecasts). The results and the analysis tools developed in this project have improved the understanding of low-latitude upper-level disturbances and their impacts. The project has also advanced the verification of ensemble predictions. Both can contribute to an improvement of numerical weather forecasting in the long term.

Projektbezogene Publikationen (Auswahl)

 
 

Zusatzinformationen

Textvergrößerung und Kontrastanpassung