Klemp (1987) identifies two factors that determine mid-latitude supercell motion: 1. Advection of the storm by the mean wind 2. Motion resulting from interaction of a convective updraft with the vertically-sheared environment Mean supercells motions are up to 10 m/s to the right of the mean wind (MW) along a line roughly perpendicular to the vertical wind shear (SH) through a deep layer. The rightward deviant motion is largely due to favorable dynamic vertical pressure gradient forcing produced on the storm flanks as a result of updraft rotation. Assuming non-negligible deviant motion Bunkers et al. (2000) developed a shear-relative and Galilean invariant method (or independent of vertical wind profile differences when the same vertical wind shear is present) for predicting mid-latitude supercell motion from environmental soundings: where: V SPC = supercell motion vector V MW = mean wind vector in the 0-6 km layer D = deviation from the mean wind of 7.5 m/s V SH = vertical wind shear computed as the vector difference between the 5.5-6 km layer mean and the 0-0.5 km layer mean We apply the same logic and estimate TC supercell motions for a wide range of V MW , D, and V SH from each proximity sounding • V MW ranged from 0-3 km to 0-9 km → 0-6 km optimal • D ranged from 0-8 m/s → 3.0 m/s optimal • V SH top ranged from 4-8 km → 5.5-6 km optimal  The selected optimal combination (SPC-OPT) minimized the MAE between the observed and forecast motion vectors  The mean SPC-OPT is consistent with the mean observed supercell motion (Ο) and exhibits statistically insignificant error with the smallest KD area centered on “zero error”  Use of SPC-OPT in CRH calculations result in 50-100 m 2 /s 2 lower magnitudes than those estimated using the mid-latitude supercell motion predictions. Data and Methods First, the NHC-HURDAT database was used to identify all U.S. landfalling TCs from 1997–2008. Second, the SPC-ONETOR database was searched to identify all tornadoes reported < 800 km of a TC center. A cursory examination of available surface analyses was performed to ensure tornadoes were embedded within the TC circulation. Third, the ESRL-RAOB database was searched to identify all rawinsondes launched within 3 h and 111 km of a TC tornado. Lastly, the NEXRAD WSR-88D archive was used to identify which tornadic cells were supercells – following the classification guidelines of Edwards et al. (2012) – located within 111 km of the radar. Such procedures identified 52 supercells with a nearly coincident/collocated “proximity” sounding. For each supercell-sounding pair the following analysis was conducted: 1. Observed supercell motions were obtained by tracking the hook echo and/or mesovortex signature in animated 0.5º radar imagery over a 30-min window centered on the tornado report time. 2. Observed non-supercell motions were obtained by tracking the reflectivity centroid of an adjacent cell over the same time window. 3. Eight (8) mid-latitude sounding-based supercell motions were computed based on methods outlined in the following studies: a. Maddox (1976) – M76 b. McCaul (1991) – M91 c. Davies and Johns (1993) – DJ93 d. Davies (1998) – D98 e. Rasmussen and Blanchard (1998) – RB98 f. Bunkers et al. (2000) – BUNK g. modified RB98 – Ramsay and Doswell (2005) – RB98mod h. modified BUNK – Ramsay and Doswell (2005) – BUNKmod 4. All winds and motion vectors were placed in the cylindrical framework (azimuth/radius) with radial (VR) and tangential (VT) velocity components (but the TC motion was not removed) Sounding-based Prediction of Supercell Motions in Tropical Cyclones Matthew D. Eastin and Cameron Self Department of Geography and Earth Sciences, University of North Carolina at Charlotte Motivation and Objectives Landfalling tropical cyclones (TCs) regularly spawn tornadoes as the outer rainbands move onshore (Edwards 2012). Many tornadoes are spawned by “miniature supercells”, which are shallower, less intense, and shorter-lived than their mid-latitude counterparts (Eastin and Link 2009). Moreover, application of mid-latitude tornado forecasting techniques to TCs have shown limited success, perhaps in part, due to significant differences between the parent supercells. For example, sounding-based forecast parameters, such as cell-relative helicity (CRH; see Rasmussen and Blanchard 1998), require accurate predictions of supercell motion in order to determine which TC regions contain above- average CRH, and thus are more likely to spawn tornadoes. Hence, incorrect predictions of cell motion could produce large CRH errors, which, in turn, could significantly influence forecast/warning decisions. For mid-latitude supercells, several sounding-based methods of predicting cell motion have been developed and tested (see Ramsay and Doswell 2005). For TC supercells, no prediction method has been developed, and no feasibility study evaluating the application of the mid-latitude methods in TCs has been conducted. Nevertheless, the mid-latitude methods have been applied in TCs (Baker et al. 2009; Molinari and Vollaro 2010; Edwards et al. 2012). Thus, the objectives of this study are to first evaluate the feasibility of applying the mid-latitude cell motion prediction methods to TC supercells; and second, if needed, develop a improved sounding- based method which provides accurate prediction of TC supercell motions. We anticipate that any such improved prediction method will enhance the situational awareness for those severe weather forecasts unique to tropical cyclones. Tropical Cyclone (TC) Supercell Database Evaluation of Mid-latitude Methods Applied to TC Supercells Development of a TC Supercell Motion Prediction Method References and Additional Reading Baker, A. K., M. D. Parker, and M. D. Eastin, 2009: Environmental ingredients for supercells and tornadoes within Hurricane Ivan. Weather and Forecasting, 24, 223-244. Bunkers, M. J., B. A. Klimowski, J. W. Zeitler, R. L. Thompson, and M. L. Weisman, 2000: Predicting supercell motion using a new hodograph technique. Weather and Forecasting, 15, 61-79. Davies, J. M., 1998: On supercell motion in weaker wind environments. Preprints, 19 th Conference on Severe Local Storms, Minneapolis, MN, American Meteorological Society, 685-688. Davies, J. M., and R. H. Johns, 1993: Some wind and instability parameters associated with strong and violent tornadoes. Part I: Wind shear and helicity. The Tornado: Its Structure, Dynamics, Prediction, and Hazards, Geophysical Monographs, No. 79, American Geophysical Union, 573-582. Eastin, M. D., and M. C. Link, 2009: Miniature supercells in an offshore outer rainband of Hurricane Ivan (2004). Monthly Weather Review, 137, 2081-2104. Edwards, R, 2012: Tropical cyclone tornadoes: A review of knowledge in research and prediction. Electronic Journal of Severe Storms Meteorology, 7, 1-61. Edwards, R., A. R. Dean, R. L. Thompson, and B T. Smith, 2012: Convective modes for significant severe thunderstorms in the contiguous Unites States. Part III: Tropical cyclone tornadoes. Weather and Forecasting, 27, 1507-1519. Klemp, J. B., 1987: Dynamics of tornadic thunderstorms. Annual Review of Fluid Mechanics, 19, 369-402. Maddox, R. A., 1976: An evaluation of tornado proximity wind and stability data. Monthly Weather Review, 104, 133-142. McCaul, E. W. Jr., 1991: Buoyancy and shear characteristics of hurricane-tornado environments. Monthly Weather Review, 119, 1954-1978. Molinari, J., and D. Vollaro, 2010: Distribution of helicity, CAPE, and shear in tropical cyclones. Journal of the Atmospheric Sciences, 67, 274-284. Rasmussen, E. N., and D. O. Blanchard, 1998: A baseline climatology of sounding-derived supercell and tornado forecast parameters. Weather and Forecasting, 13, 1148-1164. Ramsey, H. A., and C. A. Doswell, 2005: A sensitivity study of hodograph-based methods for estimating supercell motion. Weather and Forecasting, 20, 954-970 Future Work 1. Stratify the supercells and proximity soundings with regard to their relative time and location (before/after and upwind/downwind). 2. Apply the new forecast method to a larger climatological analysis of sounding-based, cell-relative, forecast/diagnostic parameters frequently used to anticipate convective mode and severe weather. 3. Use the new forecast method in the development of an observation-based Tropical Cyclone Tornado Parameter (TCTP) for use in discriminating between environments conducive to TC supercell/tornado formation and environments not conducive. 4. Perform a similar evaluation using a larger database of TC supercells with proximity soundings obtained from the 40-km Rapid Update Cycle (RUC) numerical model analyses         × + = SH SH MW SPC V k V D V V ˆ • Total of 52 tornadic supercells from 27 landfalling TCs • All exhibited either a hook echo and/or mid-level mesovortex • All range < 400 km from the coastline between LA and VA • Majority (70%) range 200-500 km from the TC center • Majority (96%) are located east or northeast of the TC center • Majority (72%) are located < 60 km from their sounding • Distributions of associated TC and tornado intensities at the time of supercell observation are consistent with their respective climatological distributions • Sample representative of a larger TC supercell population Sounding-relative Locations TC-relative Locations Supercell Locations Supercell near MFL Hurricane Gustav 31 August 2008 0-1 km CRH 0-1 km CRH Errors Variable Deviation (D) Variable Mean-Wind (MW) Layer Variable Vertical Shear (SH) Layer Variable VT Fraction ( ) ∑       ∂ ∂ × − • − = h Sound SPC Sound z V V V k CRH 0 ˆ • Composite hodograph exhibits strong shear in the lowest 2-km and large curvature in lowest 6-km • Mean observed supercell motion (Ο) is 2-3 m/s slower than and 1-2 m/s to the right of the mean non-supercell motion ( ), and the differences are statistically significant at the 95% level • All mean forecast supercell motions (color ed ▼) are 2-8 m/s slower than and 2-6 m/s right of the the mean observed motion (Ο), and differences are statistically significant at the 95% level • Kernel Density (KD) contours (at 75%) of forecast errors show that the M91 method performs the best (smallest area; most centered on the Ο) • An improved forecast method for TCs is needed! • The M91 (■) forecast motion is consistent with the mean observed motion for non-supercells View publication stats View publication stats