Federal Department of Home Affairs FDHA Federal Office

Federal Department of Home Affairs FDHA Federal Office of Meteorology and Climatology Meteo. Swiss WG 4: interpretation and applications A success story… to be continued… Pierre Eckert Meteo. Swiss, Geneva

Topics • FIELDEXTRA presentation by JM Bettems • Retrospective (short) • Perspectives COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 2

WG 4 retrospective • 2001: Inventory of postprocessing methods, COSMO LEPS • ~2008: separation of ensemble part to WG 7 • 2009: Project INTERP (scale dependent verification), inclusion of FIELDEXTRA as universal postprocessing method • 2012: Project CORSO (winter Olympics SOCHI) Participation into various activities: • Verification and guidelines forecasters • NWP Test suite • Feed back from forecasters • Specific applications (CAT, TS indices, …) COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 3

COSMO LEPS • • • Became operational on ECMWF units Available to all COSMO states 10 km 7 km 16 20 members Extension of the domain Still used, also as a testbed / benchmark COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 4

WG 4 retrospective • 2001: Inventory of postprocessing methods, COSMO LEPS • ~2008: separation of the ensemble part to WG 7 • 2009: Project INTERP (scale dependent verification) • Inclusion of FIELDEXTRA as universal postprocessing method • 2012: Project CORSO (winter Olympics SOCHI) Participation into various activities: • Verification and guidelines forecasters • NWP Test suite • Feed back from forecasters • Specific applications (CAT, TS indices, …) COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 5

Motivation: precipitation pattern 2 km 7 km COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 6

Expected behaviour of scores From Nigel Roberts (2005) COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 7

What verification is useful to forecasters? • Verification has to be local (not too much)… • …and stratified • It can be stratified by type of phenomena (thunderstorms, drizzle, …) either observed or forecasted • Or by type of synoptic situation • The verification results can be used in order to implement targeted postprocessing (decision trees)… • …or be used by the forecasters in order • to correct the models • not to use them in certain situations • use different indicators from models (instability, …) • The verification should be relevant to the present version of the model COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 8

Work already done Precipitation bias by weather class (Zala) 2008, SW COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 2008, flat 9

Work already done Fuzzy verification by weather class (Weusthoff) We think about how to communicate the results COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 10

NE S (11 x) (10 x) YEAR 2010 differences in Fractions Skill Score for weather-type dependant verif COSMO-2 minus COSMO-7 better COSMO-2 better F N SW (18 x) (49 x) NW E H (38 x) (4 x) (73 x) SE W (4 x) (56 x) COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch (78 x) L (25 x) 11

Summary neighbourhood verification precipitation in 2010 • The skill of the models varies for different weather types and the differences between COSMO-2 and COSMO-7 varies also: - best skill: Autumn and Spring, south to northwest weather types - greatest difference COSMO-2 minus COSMO-7: Summer and Winter, north- and east types, convective cases COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch Tanja Weusthoff 12

WG 4 retrospective • 2001: Inventory of postprocessing methods, COSMO LEPS • ~2008: separation of the ensemble part to WG 7 • 2009: Project INTERP (scale dependent verification) • Inclusion of FIELDEXTRA as universal postprocessing method • 2012: Project CORSO (winter Olympics SOCHI) Participation into various activities: • Verification and guidelines forecasters • NWP Test suite • Feed back from forecasters • Specific applications (CAT, TS indices, …) COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 13

WG 4 retrospective • 2001: Inventory of postprocessing methods, COSMO LEPS • ~2008: separation of the ensemble part to WG 7 • 2009: Project INTERP (scale dependent verification) • Inclusion of FIELDEXTRA as universal postprocessing method • 2012: Project CORSO (winter Olympics SOCHI) Participation into various activities: • Verification and guidelines forecasters • NWP Test suite • Feed back from forecasters • Specific applications (CAT, TS indices, …) COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 14

WG 4 retrospective Always keep in view the bridge(s) between modellers and users (forecasters). Difficulties • The exchange of application and methods between countries is difficult. • Postprocessing and forecasters usually sit in other departments/divisions and do not participate to COSMO activities/meetings. COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 15

WG 4 perspectives • Discussions since January with SMC and STC • Nominate a focal point to WG 4 in each country • • Germany : M. Paulat Switzerland : D. Cattani Italy : A Canessa Greece : D. Boucouvala Poland : A. Mazur Romania : A. Iriza Russia : A. Bundel Israel : A. Savir, E. Brainin • STC nominates a coordinator among them • Welcome and good luck Anastasia COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 16

WG 4 perspectives Define terms of reference for WG 4 with (new) orientations. Will be precised by the WG. • Survey the development of “generic” “products” (Fog, Thunderstorms, Wind at a specific location and height, clear air turbulence (CAT), Icing, high impact weather). • “generic” means among others “not comercial”, “can be exchanged/used by all COSMO members” • Develop optimal aggregation and calibration methods • Survey (coordinate) methods intended to blend model output and observations, for instance in very short range forecasting. • Develop and promote guidelines to users/forecasters COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 17

WG 4 perspectives WG 4 should not only be considered as considering activities happening downstream of the models, but should also convey back the requirements and worries of the users / forecasters to all WGs and the COSMO management. PTs and PPs will be defined. The whole WG will meet when the survey of requirements will be complete. COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 18

Many thanks to all of you It has been my great pleasure to participate to COSMO since 2001. The consortium has since then doubled in number of member states and probably also in the amount of participating scientists. The structure and governance has evolved a lot, I hope in a sense of efficiency. The models have evolved from a 14 km resolution to 1 km with explicit convection. I would like to thank the successive SPMs Günther Doms, Tiziana Paccagnella, Marco Arpagaus, Michał Ziemiański and Dmitrii Mironov for fruitful collaboration. Also all members of the STC for the organisation and support. Last but not least all other WG coordinators for the collaboration they offered in the SMC. The relationship between forecasters and modelers is usually impregnated alternatively by hate and love. I was trying to promote the comprehension between these two worlds not only by explaining the strengths and weaknesses of the models, but also to bring back the requirements of the forecasters to the (more and more numerous) modelers. I also had the opportunity to meet interesting people from other countries. I think I can consider COSMO as a part of my (extended) family. I certainly will continue to consider these 16 years as a beautiful part of my life. COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 19

Welcome to the new life COSMO General meeting ¦ Jerusalem, September 2017 Pierre. Eckert[at]meteoswiss. ch 20