Are current sensitivity products suï¬ciently informative in targeting campaigns? A DTS-MEDEX-2009 case study

• Published in: Quarterly journal of the Royal Meteorological Society Vol. 140; no. 679; p. 525
• Main Authors: Garcies, L, Homar, V
• Format: Journal Article
• Language: English
• Published: Reading Wiley Subscription Services, Inc 01-01-2014
• ISSN: 0035-9009; 1477-870X
• DOI: 10.1002/qj.2148

The DTS-MEDEX-2009 campaign was a field experiment in which extra observations were adaptively deployed to improve the short-range forecast of Mediterranean high-impact weather (HIW) during autumn 2009. For the DTS-MEDEX-2009 cases, five dierent sensitivity analysis techniques were carried out to provide targeting guidance: singular vectors (SV) from the ECMWF; ensemble transform Kalman filter (ETKF) and Kalman filter sensitivity (KFS) from Météo France; and ensemble and adjoint sensitivities from the University of the Balearic Islands. However, the value of the targeting guidance provided by such a variety of sensitivity products has never been assessed for a Mediterranean HIW event. Since radiosonde and AMDAR profiles were the only observational means available during the DTS-MEDEX targeting campaign, this study tests the ability of each sensitivity product in identifying the region where a plausible sounding leads to a greater impact on the forecast of a potential high-impact cyclone over southern Italy on 5 December 2009. All targetable radio-sounding sites are also tested and a severe-weather meteorologist is used as a confronting reference. The verification testbed comprehends single sounding experiments and multiple sounding strategies by using the WRF Data Assimilation system. Single sounding tests reveal that sensitivity products fail to recognize the best location for a single observation since most of the soundings added over operational radio-sounding stations have a larger influence on intense cyclone forecast than the points highlighted by the objective sensitivity calculation methods. Additionally, it is shown that human-based decisions, after evaluating available sensitivity information, are not optimal, either in single or in multiple sounding strategies. [PUBLICATION ABSTRACT]