Project

Back to overview

An object-based evaluation of the forecast uncertainty of Mediterranean cyclones and their associated surface weather extremes

Applicant Wernli Heini
Number 205419
Funding scheme COST (European Cooperation in Science and Technology)
Research institution Institut für Atmosphäre und Klima ETH Zürich
Institution of higher education ETH Zurich - ETHZ
Main discipline Meteorology
Start/End 01.02.2022 - 31.01.2026
Approved amount 317'004.00
Show all

All Disciplines (2)

Discipline
Meteorology
Climatology. Atmospherical Chemistry, Aeronomy

Keywords (9)

extreme weather events; Mediterranean cyclones; atmospheric dynamics; numerical weather prediction; meteorology; forecast uncertainty; warm conveyor belts; heavy precipitation events; dust storms

Lay Summary (German)

Lead
Tiefdruckgebiete im Mittelmeerraum führen oft zu Extremwetter (Starkniederschläge, Überflutungen, Windstürme, Staubstürme). Die Prozesse, die zu diesen Tiefdruckgebieten führen und ihre Auswirkungen werden deshalb aktuell in der COST Action CA19109 untersucht. Mit diesem Projekt möchten wir dazu beitragen, die Vorhersagbarkeit von Tiefdruckgebieten im Mittelmeerraum und des mit ihnen verbundenen Extremwetters zu quantifizieren.
Lay summary

Inhalt und Ziel des Forschungsprojekts

Unser übergeordnetes Ziel ist es, zu quantifizieren, wie gut die aktuellen operationellen Ensemble-Wettervorhersagesysteme in der Lage sind, Tiefdruckgebiete im Mittelmeerraum und das mit ihnen verbundene Extremwetter zu prognostizieren. Insbesondere möchten wir verstehen, welche Aspekte für eine gute Vorhersage und eine kleine Vorhersageunsicherheit besonders wichtig sind: Ist es die dynamische Entwicklung des Tiefdruckgebiets selbst oder sind es Prozesse, die stromaufwärts über dem Nordatlantik ablaufen? Und unter welchen Bedingungen sind die Vorhersagen zuverlässig und wann sind sie es nicht? Der Fokus der Arbeit liegt beim operationellen Vorhersagesystems des ECMWF; wir werden jedoch explorativ auch erste globale konvektions-auflösende Simulationen mit einer Maschenweite von 1-3 km in die Untersuchungen einbeziehen.

Wissenschaftlicher und gesellschaftlicher Kontext des Forschungsprojekts

Unsere Arbeit wird neue und wichtige Informationen zu den physikalischen und dynamischen Prozessen liefern, die für korrekte Vorhersagen des Wettergeschehens im sehr dicht bevölkerten Mittelmeergebiet essenziell sind. Die Ergebnisse werden sowohl für die Grundlagenwissenschaft relevant sein, wie auch für die operationelle Wettervorhersage wichtige Hinweise liefern.

Direct link to Lay Summary Last update: 17.01.2022

Responsible applicant and co-applicants

Employees

Project partner

Associated projects

Number Title Start Funding scheme
185049 Understanding the influence of warm conveyor belts in extratropical cyclones on tropopause dynamics: analysis with observations, reanalyses and global climate simulations 01.01.2020 Project funding

Abstract

Mediterranean cyclones are potentially rapidly evolving weather systems, responsible for the grand majority of extreme surface conditions in densely populated Mediterranean coastal areas, includ-ing heavy precipitation and flash floods, windstorms, outbreaks of Saharan dust, and strong ocean wave events. They therefore are at the core of the recently funded COST action CA19109 “European network for Mediterranean cyclones in weather and climate”. Research in recent years led to a substantial progress in understanding the structure and evolution of Mediterranean cy-clones, their relationship to upstream processes over the North Atlantic, and their role in the formation of extreme precipitation and winds. It is therefore timely to use this increased process understanding to systematically assess the quality of numerical weather predictions of Mediter-ranean cyclones and their surface extremes. This SNF project aims at making a substantial con-tribution towards this goal, by developing an object-based approach for evaluating the state-of-the-art ensemble prediction system of the European Centre for Medium-Range Weather Forecasts (ECMWF). The novelty of this approach is its ability to elucidate the complex relationship between forecast uncertainty and the dynamical process chain from upstream conditions to Mediterrane-an cyclones to surface weather extremes, as outlined below. Thereby, this project contributes es-sentially to key objectives of the COST action and profits from the broad COST network of re-searchers, operational forecasters and stakeholders.The approach is based on the identification of two-dimensional objects of Mediterranean cyclones and other key flow features, and of surface areas with extreme precipitation, windspeed, ocean waves, and potential dust. All these objects are characterized by their location, size, and intensi-ty. They can be linked to each other in different forecasts to quantify, for instance, how strongly an error in the forecast of cyclone intensity impacts the position and/or intensity of an extreme surface weather object. More specifically, the project is structured in five parts: (1) Investigation of the ECMWF forecast performance in predicting historic Mediterranean cyclones in the last dec-ades. (2) Detailed analysis of the uncertainty in Mediterranean cyclone and extreme event objects in the current ECMWF ensemble prediction system. This key part of the project provides novel information about important questions like “Do cyclones associated with extreme events have a particularly low predictability (i.e., high forecast uncertainty)?” and “Is the position or the intensi-ty of cyclones more relevant for accurately predicting their associated surface extremes?” (3,4) Sys-tematic linkage of forecast uncertainties of Mediterranean cyclones and their extremes to the rep-resentation of upstream processes over the North Atlantic. This will address whether Mediterra-nean cyclones with highly uncertain predictions are associated with specific upstream processes (e.g., a particularly strong warm conveyor belt). (5) Assessment of the performance of prototype versions of kilometer-scale global forecasts models. We will investigate whether these high-resolution models, with a better representation of orographic effects and deep convection, lead to better predictions of Mediterranean cyclones and their surface impacts. The results of this project will be regularly discussed with representatives from all three COST action working groups.The approach is based on the identification of two-dimensional objects of Mediterranean cy-clones and other key flow features, and of surface areas with intense or extreme precipitation, windspeed, potential dust mobilization, or ocean waves. All these objects are characterized by measures of their location, size, and intensity. They can then be linked to each other in differ-ent forecasts in order to quantify, for instance, how strongly an error in the forecast of cyclone intensity impacts the position and/or intensity of an extreme surface weather object. More spe-cifically, the project is structured in four parts: (1) Investigation of the ECMWF forecast perfor-mance in predicting intense historic Mediterranean cyclones and their associated extreme events in the period from 1979 until today. This will elucidate how strongly the operational forecast quality improved during the last decades. (2) Detailed analysis of the uncertainty in Mediterranean cyclone and extreme event objects in the current ECMWF ensemble prediction system. This key part of the project provides novel information about important questions like “Do cyclones associated with extreme events have a particularly low predictability (i.e., high forecast uncertainty)?” and “Is the position or intensity of cyclones more relevant for accurately predicting their associated surface extremes?” (3) Systematic linkage of forecast uncertainties in the Mediterranean to the representation of upstream processes over the North Atlantic. This will address whether Mediterranean cyclones with highly uncertain predictions are associated with specific upstream processes (e.g., a particularly strong warm conveyor belt). (4) Assessment of the performance of prototype versions of convection-permitting global forecasts models. We will investigate whether these high-resolution models, with a better representation of orographic effects and deep convection, lead to better predictions of Mediterranean cyclones and their sur-face impacts. The results of this project will be constantly discussed with representatives from all three COST action working groups.
-