PhD STUDENT: Angela Cocozza Tutor: Ch.mo Prof. Piero Lionello Co-tutor: Ch.mo Prof. U. Ulbrich

SCIENCES OF CLIMATE CHANGES XXV CYCLE “Intercomparison of mid latitude storm diagnostics”

Contents

• Background / aims of the project• methods• First results of project• Conclusions

IMILASTIntercomparison of mid

latitude storm diagnostics

Diagnostics of the observed and projection of the future changes of extratropical storms are a key issue e.g. for insurance companies, risk management and adaptation planning. Storm-associated damages are amongst the highest losses due to natural disasters in the mid-latitudes. Therefore the knowledge of the future variability and change in extratropical cyclone frequency, intensity and track locations is crucial for the strategic planning and minimization of the disaster impacts. Characteristics of cyclone activity may seriously depend on the methodology of cyclone identification and tracking. Linear trend magnitude and even sign might depend on the detection and tracking methods of the cyclones. Therefore, a special effort is needed to perform an intercomparison and to quantify these uncertainties. A proper knowledge about advantages and restrictions of different schemes must be obtained to be able to provide a synthesis of results rather than puzzling the scientific and the general public with apparently contradicing statements.

Project

Project team:

• S. Gulev, N. Tilinina, I. Rudeva, P.P. Shirshov Inst. of Oceanology, Moscow/RUS

• J. Pinto, S. Ulbrich, University of Köln/GER

• M. Sinclair, Embry-Riddle Aeronautical University, Prescott, Arizona/USA

• O. Zolina, University of Bonn/GER

• C. Schwierz, ETH Zurich/SUI

• X.L. Wang, Y. Feng, Environment Canada/CAN

• U. Ulbrich, G. Leckebusch, F-Univ. Berlin/GER

• R. Blender, University of Hamburg/GER

• C. Raible, University of Bern/SUI

• K. Hodges, H. Dacre, S. Gray, University of Reading/UK

• R. Benestad, Ina Kindem, Met Office, Oslo/NOR

M. Inatsu, Hokkaido Univeristy/JAPI. Simmonds, K. Keay, University of

Melbourne/AUSH. Wernli, M. Sprenger, ETH Zürich/SUII. Trigo, University of Lisbon/PORR. Caballero, University of

Stockholm/SWEJ. Hanley, Univeristy of Dublin/IRLT. Hewson, ECMWF/UKS. Kew, KNMI/NEDP. Lionello, A. Cocozza, University

of Salento, Lecce/ITAH. von Storch, L. Xia, GKSS

Geesthacht/GERM. Akperov, I. Mokhov, A.M. Obukhov

Inst. of Atmos. Physics, Moscow/RUS

M. Liberato, Universidade de Trás-os-Montes e Alto Douro/POR

Executive committee:U. Neu, ProClim, Bern/SUI (project coordination; sponsored by Swiss Re)X.L. Wang, U. Ulbrich, G. Leckebusch, C. Raible

Project

Background

What is the project about?• Characteristics of cyclone activity and quantification of trends

depend on the methodologies used for storm track detection• We need knowledge about:

advantages and limitations of different analysis schemes and their influence on results

better synthesis of results and proper interpretations

● Aims of the project● Intercomparison of the results using different analysis methods● Assessment of the methodologies● point out the specific information given by specific methods

Project

What did we do?

● Evaluated different methods

differences in schemes

● Defined a standard intercomparison experiment, calculating storm tracks with all individual methods, using

● ERA-interim 1.5° reanalysis 1989 - 2009 ● - ERA-interim 0.75° reanalysis for individual storms

● differences and similarities in results

Project

Differences in schemes

● Pre-processing:- grid transformation ( different resolution)- data filtering

● Identification method:- different metrics (SLP minima; vorticity; pressure contours;

850 hPa minima)- different elimination criteria (for vorticity, core pressure,

pressure gradient, amplitude, terrain height, lifetime)- different treatment of special cases (splitting of cyclones; inclusion/exclusion of open systems; cyclone distance)

● Tracking

● Intensity measures (vorticity, SLP minima, gradient)

● Post-processing

Project

Differences - ExamplesUse of different metrics

Example of storm "KLAUS"

M08 (based on SLP minima)vs.M02 (based on vorticity)

Liberato et al.

Rudeva and Gulev

All cyclonesDetected by

ξ850 but missed in SLP

Difference between cyclones detected by 850hPa vorticity and SLP (M12)

Vorticity detects cyclones earlier and weaker ones

Project

Differences - Examples

● Inclusion/exclusion of open structures

opensystem

TT

closedcontours

M11: 1181 closed / 986 open depr. M16: 1630

M11: 429 closed / 301 open depr.M16: 470

M11: 124 closed / 62 open depr.M16: 117

Reale, Lionello et al.

Cyclogenesis in the Mediterranean:

All

EM

WM

Project

Methods

IDENTIFICATION AND TRACKING OF CYCLONES

The method carries out the partitioning of the SLP fields in depressions by the identification of sets of steepest descent paths leading to the same SLP minimum.

Each grid point is connected to the lowest of the 8 nearest-neighbor grid points. This step is repeated until a pressure minimum, which is, obviously, a point where the SLP value is lower than the SLP at the 8 nearest grid points, is reached. All the points crossed by a path leading to the same minimum are assigned to the same depression.

Publications using this method:•Lionello P. and F. Giorgi , (2007), Winter precipitation and cyclones in the Mediterranean Region: future climate scenarios in a regional simulation Adv. Geosci. , 12, 153-158 •Lionello P., U.Boldrin, F.Giorgi, (2007), Future changes in cyclone climatology over Europe as inferred from a regional climate simulation Clim.Dyn DOI 10.1007/s00382-007-0315-0 •De Zolt S., P.Lionello, P.Malguzzi, A.Nuhu, A.Tomasin (2006) The disastrous storm of 4 November 1966 on Italy Nat. Hazards Earth Syst. Sci., 6, 861-879 •Lionello P, F.Dalan, E.Elvini (2002) Cyclones in the Mediterranean Region: the present and the doubled CO2 climate scenarios, Clim. Res.,22, 147-159

The algortims reads MSLP maps and returnsA list of low-pressure systemsSLP minimum, extension , max vorticity, max gradient, depth for each systemTrack of the SLP minimum and of the vorticity maximum

ProgramLoop on the maps

read mapfind the minima (assign markers to each

grid point) merge minima (exclude from the list

small and shallow minima)for each minima compute diagnostic

(extension, max vort, max gradient, depth…)

compare position of minima between successive maps and identify tracks

output a track when cyclolisis occursEnd of loop

Call vortgrad

Call fndcycl

call_curve

Methods

IDENTIFICATION AND TRACKING OF CYCLONES

SLP FIELD

LOCAL MINIMA

LOW PRESSURE SYSTEMS

Fig. 1. Example of the procedure for the identification of the cyclones. (a) Original sea-level pressure (SLP) field. (b) Results of the partitioning procedure. Each dot represents a grid point, and the dots with the same gray level belong to the same partition. Black dots show the location of the pressure minimum of each partition. (c) Final set of large depressions that result from the merging of the small depressions whose central minimum is at a distance less than 4 grid points from the boundary of a different and deeper depression.

Based on P. Lionello et Al. Clim Res Vol. 22: 147–159, 2002

Methods

Range for merging of low pressure systems into single cyclones= 5/gridstep

Actual tracking (example in the following slides) is carried out using for each system its baricentrum, not the position of the minimum

Baricentrum= average position of grid point belonging to the cyclone within a 3hPa range above the minimum

Methods

Methods

Methods

Methods

Methods

The first resultsInput data sets:1. ERA interim, 1.5°/6h, 01-Jan-1989 – 31-Mar-20092. ERA interim, 0.75°/6h, 01-Jan-1990 – 31-Mar-19903. ERA interim, 0.75°/6h, specific storms:a) Daria 25-01-1990, Germany 19/1/90 00UTC - 28/1/90 18UTCb) Vivian 26-02-1990, Germany 20/2/90 00UTC - 29/2/90 18UTCc) Dec92 9-12-1992, AquaAlta 3/12/92 00UTC - 12/12/92 18UTCd) Lili 28-10-96, UK 22/10/96 00UTC - 31/10/96 18UTCe) Yuma 24-12-97, UK 18/12/97 00UTC - 27/12/97 18UTCf) Anatol 3-12-99, Denmark 27/11/99 00UTC - 6/12/99 18UTCg) Lothar 26-12-99, France 20/12/99 00UTC - 29/12/99 18UTCh) Martin 28-12-99, France 22/12/99 00UTC - 31/12/99 18UTCi) Oratio 30-10-00, UK 24/10/00 00UTC - 2/11/00 18UTCj) Algeria 11-11-2001, Algeria 5/11/01 00UTC - 14/11/01 18UTCk) Jeanette 27-10-02, Germany 21/10/02 00UTC - 30/10/02 18UTCl) Slovakia 29 Nov 04, Slovakia 23/11/04 00UTC - 2/12/04 18UTCm) Dagmar 17-12-04, France 11/12/04 00UTC - 20/12/04 18UTCn) Erwin 08-01-05, Sweden 2/1/05 00UTC - 11/1/05 18UTCo) Gordon 20-09-06, Ireland 14/9/06 00UTC - 23/9/06 18UTCp) Renate 03-10-06, France 27/9/06 00UTC - 6/10/06 18UTCq) Kyrill 18-01-07, Germany 12/1/07 00UTC - 21/1/07 18UTCr) Paula 25-01-08, N Europe 19/1/08 00UTC - 28/1/08 18UTCs) Klaus 24-01-09, France 18/1/09 00UTC - 27/1/09 18UTC

Results

Algeria 11-11-2001 Algeria Slovakia 19-11-2004 Slovakia

Erwin 08-01-2005 Sweden

Renate 03-10-2006 France Paula 25-01-2008 N Europe

Results

Daria 25-01-1990 Germany

• Fare clic per modificare stili del testo dello schema– Secondo livello– Terzo livello

• Quarto livello– Quinto livello

Vivian 26-02-1990 Germany

Lili 28-10-1996 UK

Yuma 24-12-1997 UK

Anatol 3-12-1999 France

Results

Cyclone tracking in ERA-interim 1.5

Cyclone track identified using LeA (blue) and BeA (fuchsia) corresponding to winter storms reported in the IMILAST intercomparison protocol. Red dot represents the position of the cyclone in the day of its maximum intensity

The tracks produced by LeA and BeA are very similar, particularly during the stage when cyclones reach maximum intensity ( red dot )

LeA produces generally a longer track than BeA because of problems with the identification of the system during the weak stages of its lifetime .

A preliminary intercomparison between the tracking algorithms of Freie Universität Berlin and Università del Salento . M.Reale, K.Nissen, P.Lionello, U.Ulbrich

Results

in conclusion work in progress…

● Evaluation of differences between methods, (e.g. numbers, geographical distribution of tracks)

• an intercomparison of cyclone tracking methods in relation with extreme events in the Mediterranean region

Activities Description Acquired CFU DateSeminar attended

5 CFU

Impacts of climate changes over Apulia

Certificate of attendance (Yes)

Lecce 15 November 2010

Formative activities approved by the PhD council

5 CFU

International MedCLIVAR-ICTP-ENEA Summer School

on the Mediterranean Climate System and Regional

Climate Change

Certificate of attendance

(Yes)

Miramare , ICTP Trieste 13 -22 September

2010

Attendance at the “MedClivar FINAL conference-Mediterranean climate from past to future”

Certificate of Attendance (Yes)

Lecce 6-9 June 2011

Activities connected with the research 5 0 CFU

Workshop Intercomparison of

Mid-latitudeStorm diagnostics

presentation of the first results

Interlaken, Swiss 30/3 – 1/4

The IMILAST project : Influence of including open cyclones in storm identification schemesMarco Reale , Piero Lionello , Angela Cocozza , and IMILAST team

WCRP OSC Climate Research in Service to Society24-28 October 2011Sheraton Denver Downtown HotelDenver, CO, USA

An intercomparison of cyclone tracking methods in relation with extreme events in the Mediterranean regionA. Cocozza, M. Reale, P. Lionello, and IMILAST team

11th EMS Annual Meeting10th European Conference on Applications of Meteorology (ECAM)12 – 16 September 2011Berlin, Germany

THANK YOU FOR ATTENTION