11C.1 The Impact of Surface Heat Fluxes outside the Inner Core on the Rapid Intensification of Typhoon Soudelor (2015) Chin-Hsuan Peng and Chun-Chieh Wu Department of Atmospheric Sciences, National Taiwan University, Taipei, Taiwan rd Conference on Hurricanes and Tropical Meteorology 33 rd April 18, 2018 @ @ 33 Ponte Vedra, Florida, United States Acknowledgments : Grant: Ministry of Science and Technology (Taiwan)
Introduction – TC intensity change and rapid intensification (RI) Challenge of TC intensity forecast • The forecast skill of TC intensity is regarded to be a challenging topic (NHC 2013; JMA 2013; Ito 2015) . • Unexpected RI episodes could cause serious loss of life and property to the coastal regions (Chang and Wu 2017) . Definition of RI • In terms of minimum central pressure (P min ) : ≥ 42 hPa / 24 hr drop (Holliday and Thompson 1979) • In terms of surface maximum tangential wind (V max ) : ≥ 35 kts / 24 hr increase (Kaplan et al. 2010; Lee et al. 2016) Favorable inner-core conditions to RI Low-level high θ e air Convective burst Upper-level warming (Montgomery et al. 2006; Barnes (Heymsfield et al. 2001; Reasor (Zhang and Chen 2012; Chen and et al. 2009; Guimond et al. 2010; and Fuentes 2010; Miyamoto and Secondary circulation Inertial stability Zhang 2013; Wang and Wang 2014) Zhang and Chen 2012; Chen and Takemi 2013; Wang and Wang 2014) Zhang 2013; Rogers et al. 2013; (e.g., Eliassen 1951; Ooyama 1969; Ooyama Wang and Wang 2014; Chen (Schubert and Hack 1982; and Gopalakrishnan 2015) 1982; Shapiro and Willoughby 1982) Vigh and Schubert 2009) 2 11C.1
Introduction – The role of surface heat fluxes in TC intensification • When the surface wind speed (U) is artificially Control capped (at ≥ 5 m/s) in the whole domain, TC still reach hurricane intensity while averaged U = 20 m/s intensification rate decrease due to the reduction of surface heat fluxes . Intensity (Montgomery et al. 2015) U = 10 m/s ◦ Surface sensible heat fluxes: SHF = ρ c p C H U( Δθ ) U = 5 m/s ◦ Surface latent heat fluxes: Time LHF = ρ L v C Q U( Δ q) (Green and Zhang 2013) (Zhang and Emanuel 2016) 3 11C.1
However, would the limitation of surface heat fluxes always lead to a reduction of TC intensification rate? 4 11C.1
SHF = ρ c p C H U ( Δθ ) Model and experimental design LHF = ρ L v C Q U ( Δ q) Model setting in WRF simulation (V3.6.1) Sensitivity experiments • The surface wind ( U ) is capped at 1 m/s . Data ERA-Interim reanalysis data (ECMWF) (It means that surface fluxes are mostly suppressed .) Two-way interactive, movable, The flux-suppressed (blue shaded) regions are as follows: Domain triply nested grid ( 15 / 5 / 1.67 km ) CTRL 10IC 15IC 20IC Vertical levels 41 eta levels (model top set at 30 hPa) 120km 90km Simulation period 7/31 12Z ~ 8/3 12Z (including RI phase) 60km 60km Microphysics WSM 6-class graupel scheme 500km Yonsei University parameterization (YSU) Boundary layer 25IC 30IC 40IC 50IC Radiation RRTMG scheme 300km 240km 180km 150km Cumulus Kain-Fritsch scheme (only for domain 1) Initialization Digital filter initialization (DFI) 5 11C.1
Start (RI-24hr) RI onset 8/2 06Z prior to RI during RI 6 11C.1
10IC 15IC Non-RI case CTRL 30IC 40IC RI case Domain size: 370 × 370km Field: reflectivity (1km height) 7 11C.1
Sensitivity experiments – Secondary circulation difference (prior to RI) 10IC - CTRL 15IC - CTRL Z CTRL 30IC - CTRL 40IC - CTRL RMW Shaded : Vertical wind ( W ) Vector : ( Vr , W × 10) R 8 11C.1
Sensitivity experiments – Secondary circulation difference (during RI) 10IC - CTRL 15IC - CTRL Z CTRL 30IC - CTRL 40IC - CTRL Shaded : Vertical wind ( W ) Vector : ( Vr , W × 10) R 9 11C.1
Sensitivity experiments – CFADs of vertical velocity within the RMW (prior to RI) 40IC - CTRL 10 11C.1
Sensitivity experiments – Inertial stability and diabatic heating (prior to RI) Z CTRL 15IC 40IC 10 10 10 20 20 20 RI case Non-RI case RI case R 11 11C.1
Sensitivity experiments – Surface fluxes & wind speed at the lowest level Time F ss = C E 𝜍 V s ( S s − S 0 ) CTRL 40IC ( Juračić and Raymond 2016) RI + 12hr RI onset RI – 12hr Shaded : Surface wind speed ( V s ) Contour : Surface fluxes of moist entropy R 12 11C.1
Sensitivity experiments – Diabatic heating-generated vorticity (inner core) Y 𝜖ζ 𝜖H D CTRL 40IC 𝜖t ∝ 𝜖z ( Juračić and Raymond 2016) ζ : Relative vorticity H D : Diabatic heating Shaded: Diabatic heating difference between 0.5 - 3km height Contour: Diabatic heating-generated vorticity averaged from 1 to 2km height X 13 11C.1
Sensitivity experiments – Instability in the lower troposphere (θe* & RH) (RI – 12hr) Z CTRL 40IC 3km height Surface Shaded: RH Contour: θ e * R 14 11C.1
Sensitivity experiments – Instability in the lower troposphere (θe* & RH) (RI onset) Z CTRL 40IC 3km height Surface Shaded: RH Contour: θ e * R 15 11C.1
Sensitivity experiments – Instability in the lower troposphere (θ e* & RH) (RI + 12hr) Z CTRL 40IC 3km height Surface 80km Shaded: RH Contour: θ e * R 16 11C.1
Conclusions – Schematic diagram 25IC 30IC Z Enhanced eyewall updraft 40IC 50IC Stronger, More compact Less active rainbands warm core Larger Suppressed V max heat fluxes ≥ 2.5 × inner-core size 11C.1
Conclusions – Flowchart Convection develops Suppressed the surface Convectively-generated PV closer to TC center fluxes outside the inner core concentrates near TC center Positive effect Negative effect Less heat energy Inertial stability ↑ transported into the inner core in the inner-core region The most violent winds Relatively dry air TC inner core gains more concentrate in the intrudes into a TC energy from the ocean inner-core region 10IC 25IC 40IC Weaker TC Stronger TC 15IC 30IC 50IC 11C.1
Ongoing & future works • Cap the surface wind at 5, 10, 20 m/s in the calculation of surface heat fluxes. • Investigate the relation between TC intensification rate and the surface heat fluxes in different radial extents. • Classify the asymmetric processes before and during RI, especially the relationship between rainbands and inner core. 11C.1
Recommend
More recommend
