Identifying mesoscale high-wind features within extratropical cyclones Lea Eisenstein, Peter Knippertz, Joaquim Pinto

Institute of and Research, Karlsruhe Institute of Technology vEGU21, Session AS1.6, 30 April EGU21-2248 Causes of high winds

• Warm conveyor belt jet (warm jet, WJ) • Cold conveyor belt jet (cold jet, CJ) • Sting jet (SJ) • (Post-)cold frontal convective gusts (CFC)

→Differ in timing within cyclone life cycle, location relative to cyclone center, size of footprint, intensity and dynamicsl characteristics Adapted from Clark and Gray (2018), Fig. 7 →Do forecast errors also differ? →Can regime-dependent post- processing improve the forecast?

Adapted from Hewson and Titley (2010) and Hewson and Neu (2015) Data

• Dataset of hourly surface observations (2001 – 2020, see upper figure): • Europe: MSLP, temperature, precipitation, wind speed and wind direction • : As above plus gust speed and dew point temperature • COSMO-DE/D2-EPS forecasts (see lower figure) • 3D grid for nine case studies between 2017 – 2020 with COSMO-DE differing developments, tracks and feature occurrences (see next slide) COSMO-D2 • interpolated to station locations for 2011 – 2019 (soon 2020) • Deterministic ICON-EU forecasts and ICON-LAM simulations for four of the nine case studies (Shapiro- Keyser cyclones)

3 Case studies: winters 2017/18 – 2019/20

Xavier Herwart Burglind Friederike Fabienne Bennet Eberhard Sabine Diana II Date 05.10.2017 29.10.2017 03.01.2018 18.01.2018 23.09.2018 04.03.2019 10.03.2019 9.-10.02.2020 1.-2.03.2020 Max. gust 176 kmh-1 176 kmh-1 217 kmh-1 204 kmh-1 158 kmh-1 144 kmh-1 166 kmh-1 176 kmh-1 132 kmh-1 in Germany Brocken Brocken Weinbiet Feldberg Feldberg Feldberg Feldberg (>500 m) (1134 m) (1213 m) (1490 m) (1134 m) (553 m) (1490 m) (1490 m) (1490 m) (1490 m) Max. gust 124 kmh-1 144 kmh-1 120 kmh-1 138 kmh-1 144 kmh-1 114 kmh-1 126 kmh-1 155 kmh-1 107 kmh-1 in Germany Holzdorf Alte Büchel Gera-Laubnitz Konstanz Arkona Alsfeld Fürstenzell Rheinstetten (<500 m) (81 m) Leuchtturm (32 (477 m) (311 m) (483 m) (183 m) (305 m) (480 m) (116 m) m) SSI 4.35 5.38 5.49 9.96 3.76 3.82 7.77 12.39 0.59 Damage 324 Mio. € 264 Mio. € 756 Mio. € 1672 Mio. € < 200 Mio. € < 200 Mio. € 782 Mio. € 1571 Mio. € < 200 Mio. €

Wind cause CJ, WJ Strong ∇p, WJ, CJ, CFC SJ, CJ, WJ WJ, CFC, CFC, WJ, CJ CJ, WJ WJ, CFC, SJ, CJ CFC weak CJ strong ∇p

4 Distinction between WJ and CFC

Example: Burglind While CFC MSLP shows heavy decreases (03 Jan. 2018, 7 UTC) precipitation, ahead of the COSMO-DE-EPS, Member 1 there is almost cold front, such init: 2018010221 none ahead of that the the front, i.e. in tendency is ≲ 0 the WJ region for the WJ while it becomes positive when CFC arrives

For CFC, θ The wind decreases direction with the onset changes with of precipitation arrival of the cold front, hence CFC produced by the front detection Contours: Gust of Met.3D speeds normalised See Session AS1.3, vPICO EGU21- by the 98th 10055 by Andreas Beckert et al. percentile of gusts

normalised by the median 5 Interactive identification

How does a person identify dynamical features with their synoptic knowledge? → Building an interactive tool to select feature areas by hand Data table of all stations with Set label of parameters and selected stations Case study set labels in the table to a specific feature Select different times Get histograms of parameters for all Choose which and selected stations variable you (light and dark blue, want to see respectively)

Select area with mouse

6 Characteristics of the identified features

CJ+SJ* • Pressure tendency CFC and precipitation are WJ distinctive characteristics as expected • Potential temperature tendency is less distinctive than expected → alternative approach needed to detect warm sector • Wind direction tendency of little use

*The distinction of CJ and SJ is not possible in surface 7 observations so that the features are displayed together Summary and future plans

• Development of a tool to interactively identify mesoscale wind features within European cyclones helped extracting their characteristics in surface observations. • Distinction between CJ and SJ from surface observations alone is very difficult and hence the two features are taken together for now. • CJ, WJ and CFC can be distinguished from precipitation, pressure and temperature fields. • Next steps – Expand interactive tool and identification to gridded data. – Use random forests for feature classification. • Planned – Produce feature climatology using observations (2001–2020) & COSMO-DE/D2-EPS forecasts. – Distinguish CJ and SJ from gridded data without explicit use of Lagrangian trajectories. – Experiment with instability-based SJ precursor tool (Gray et al., 2021) and kinematic SJ identification (Manning et al. vPICO EGU21-12795). – Feature-dependent forecast error analysis and post-processing in collaboration with mathematicians in Waves to Weather (Benedikt Schulz, Sebastian Lerch).

8 Sources

• Clark, P.A. and Gray, S.L. (2018) Sting jets in extratropical cyclones: a review. Q J R Meteorol Soc., 144: 943–969. doi: 10.1002/qj.3267 • Klawa, M. and Ulbrich, U. (2003) A model for the estimation of storm losses and the identification of severe winter storms in Germany. Nat. Hazards and Earth Syst. Sci., 3: 725–732.doi: 10.5194/nhess-3-725-2003 • Hewson, T. and Titley, H.A. (2010) Objective identification, typing and tracking of the complete life- cycles of cyclonic features at high spatial resolution. Meteorol. Appl., 17: 355-381. doi: 10.1002/met.204 • Hewson, T. and Neu, U. (2015) Cyclones, windstorms and the IMILAST project. Tellus A, 6, 1-33. doi: 10.3402/tellusa.v67.27128 • Gray, S.L., Martínez‐Alvarado, O., Ackerley, D. and Suri, D. (2021) Development of a prototype real‐time sting‐jet precursor tool for forecasters. Weather. doi: 10.1002/wea.3889 • Beckert, A., Eisenstein, L., Hewson, T., Craig, G. C., and Rautenhaus, M. (2021) Objective 3D atmospheric front detection in high-resolution numerical weather prediction data, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-10055, doi: 10.5194/egusphere-egu21-10055 • Manning, C., Kendon, E., Fowler, H., Roberts, N., Berthou, S., Suri, D., and Roberts, M. (2021) Midlatitude cyclones in convection permitting climate simulations: the added value offered for extreme wind speeds and sting-jets, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-12795, doi :10.5194/egusphere-egu21-12795

9