Impact of the Diurnal Radiation Cycle on Secondary Eyewall Formation Xiaodong Tang 1 , Zhe-Min Tan 1 , Juan Fang 1 , Y. Qiang Sun 2 , and Fuqing Zhang 2 1 School of Atmospheric Sciences, Nanjing University, China; 2 Department of Meteorology, The Pennsylvania State University, USA 16/18 16/06 15/18 14/18 15/06 UTC 16/18 16/06 15/18 14/18 15/06 UTC (a) (b) (c) (d) [dBZ] 10 m s -1 Control NoSolarRad (b) (c) 2100 UTC 16 Sep C 16 Sep T 2100 U T T C 15 17 20 22 25 27 30 32 35 37 40 42 45 47 [dBZ] Observation (a) 2008 UTC 16 Sep Control NoSolarRad [m s -1 ] [m s -1 ] (a) (d) (b) (e) (c) (f) Vt @ z = 1 km Vr @ z = 1 km Vr @ z = 2 km 17/12 17/00 16/12 16/00 15/12 15/00 14/12 UTC 17/12 17/00 16/12 16/00 15/12 15/00 14/12 UTC SEF Figure 1: Composite radar reflecvity from (a) aircraſt reconnaissance into Hurricane Edouard (2014), and (b) CNTL, and (c) NoSolarRad at a height of 5 km. Red arrows denote vercal shear vectors of averaged environmental wind. Figure 2: Hovmöller plots of azimuthal-mean tangenal velocity at a height of 1 km and radial velocity at heights of 1 and 2 km for (a)–(c) CNTL and (d)–(f) NoSolarRad. The superposed black lines denote the RMW at 1 km. Figure 3: Hovmöller plots of azimuthal mean vercal velocity at heights of 6 and 2 km for (a)–(b) CNTL and (c)–(d) NoSolarRad. The vercal red lines show key radial distances. Figure 5: Evoluon of (a) 2-m temperature, (b) MCAPE, (c) 10-m wind speed, and surface fluxes of (d) latent heat and (e) sensible heat averaged between 60 and 75 km radius for CNTL and NoSolarRad. Figure 4: Height–me plot of NoSolarRad minus CNTL di fference of (a) shortwave radiave heang, (b) longwave radiave heang, (c) net radiave heang, and (d) latent heang between 60 and 75 km radius from 1300 UTC 14 September to 1800 UTC 16 September 2014. The units are 10 –5 K/s for (a)–(c) and 10 –3 K/s for (d). • Heated surface air weakens WISHE feedback between the surface fluxes (that promote convection) and convective heating (that feeds into the secondary circulation and then the tangential wind). • The net radiative heating in CNTL is much stronger due to the solar insolation at daytime. • Less conducive for deep moist convection in CNTL • Less diabatic heating due to suppressed convection in CNTL • Difference: 0.5–1 K/day at the top of the boundary layer 4. Radia ve e ffects on SEF 4.1 Moat formaon 4.2 Balanced aspects of SEF 3. Overview of the SEF of Hurricane Edouard (2014) 3.2 Evoluon of vercal velocity (Fig. 3) 3.1 Evoluon of boundary layer wind (Fig. 2) Control NoSolarRad z = 6 km z = 2 km (a) (b) (c) (d) [m s -1 ] -1 -0.5 -0.25 -0.05 0.1 0.2 0.5 1 1.5 2 2.5 3 17/12 17/00 16/12 16/00 15/12 15/00 14/12 17/12 17/00 16/12 16/00 15/12 15/00 14/12 UTC UTC inner rainbands moat • The latent heating released from more convective activities in the inner rainbands outside of primary eyewall in NoSolarRad • The outer-core (outside the radius of 150 km) upward motion at mid-level in CNTL became more organized, and began to move inward • Clear moat formation and SEF Moat region is highly sensitive to the solar shortwave radiative heating mostly in the mid- to upper-level at daytime, which leads to a net stabilization effect and suppresses convective development. The heated surface air weakens WISHE feedback between the surface fluxes (that promote convection) and convective heating (that feeds to the secondary circulation and then the tangential wind). NoSolarRad: without solar radiation, active inner rainband, suppressed primary eyewall, no moat, no SEF The radiation-induced absence of latent heating is more important on moat formation in the early stage of SEF . 1. Introducon • Secondary eyewall formaon (SEF) is a key issue for TC research and forecas ng, as it is closely related to both short-term TC intensity change and TC size change. • We demonstrated the impacts of radiaon on the size and strength of the mature hurricane Edouard (2014) (Tang and Zhang 2016). • The mechanism by which the solar insolaon affects SEF remains unexplored, and this is the focus of the current study. 2. Experimental setup Control run (CNTL): • Real diurnal cycle; Hurricane Edouard; integrated from 1200 UTC 11-Sep-2014 Sensivity experiments: • No solar insolaon ( NoSolarRad); starng at 72 model integraon hours of CNTL The crical differences between CNTL and NoSolarRad are the stronger inner rainbands and the lack of a clear moat region in NoSolarRad. CNTL: The double maximum inflow (oulow) regions at the 1-km (2-km) level formed at ~1200 UTC 16 September (Fig. 2b, c). CNTL: The secondary maximum of the tangen al winds formed at ~1800 UTC 16 September (Fig. 2a). NoSolarRad: Heang from stronger inner rainbands outside the RMW in the midtroposphere increasing (reducing) low-level tangen al wind outside (near and inside) the RMW outward expansion of the RMW (Fig. 2d). NoSolarRad: The maximum radial inflow was farther out from the RMW during the period from 1800 UTC 15 September to 0600 UTC 16 September (Fig. 2e), without a con nuous secondary maximum of 2-km oulow (Fig. 2f). Figure 6: Radius–height cross-secons of the 2-hour azimuthal-mean of the ineral stability parameter I 2 (shading) and latent heang (blue contours, unit: 10 –3 K/s) from WRF output in (a) CNTL and (d) NoSolarRad, and of ver cal velocity (shading), radial velocity (red contours, unit: m/s), and the combined contribuon of radial advecon of absolute vorcity and vercal advec on of tangenal wind to the tangenal wind tendency (green contours, 10 –3 ms –2 ) from Sawyer-Eliassen model calculaons using (b) the background vortex structure and diabac hea ng from CNTL, (c) the background vortex structure of CNTL, diabac heang from NoSolarRad, (e) the background vortex structure and diaba c heang from NoSolarRad, and (f) the background vortex structure of NoSolarRad and diabac heang from CNTL. The period is from 0500 to 0700 UTC 16 September 2014. [m s -1 ] (c) (b) (f) vortex: CNTL latent heating: CNTL vortex: CNTL latent heating: NoSolarRad vortex: NoSolarRad latent heating: NoSolarRad vortex: NoSolarRad latent heating: CNTL (e) [10 -7 s -2 ] CNTL WRF SE WRF NoSolarRad SE SE SE (a) (d) (W m -2 ) Sensible Heat Flux (W m -2 ) (a) (b) (d) (c) (e) 16/12 16/00 15/12 15/00 14/12 UTC 16/12 16/00 15/12 15/00 14/12 UTC short wave long wave net radiation latent heat Radial wind Tangenal wind • Tang, X., and F. Zhang, 2016: Impacts of the Diurnal Radiaon Cycle on the Formaon, Intensity and Structure of Hurricane Edouard (2014), J. Atmos. Sci., 73, 2871-2892. • Tang, X. et al, 2017: Impact of the Diurnal Radiaon Cycle on Secondary Eyewall Forma on, J. Atmos. Sci., In press 6. Further reading 5. Concluding remarks