Скачать презентацию Development of a Multi-platform Satellite Tropical Cyclone Wind Скачать презентацию Development of a Multi-platform Satellite Tropical Cyclone Wind

5150d7f8defa80624241b216fd90670d.ppt

  • Количество слайдов: 24

Development of a Multi-platform Satellite Tropical Cyclone Wind Analysis System • By • John Development of a Multi-platform Satellite Tropical Cyclone Wind Analysis System • By • John Knaff, Kimberly Mueller, Mark De. Maria and Ray Zehr A demonstration project funded by NOAA Grant NA 67 RJ 0152 March 9, 2005 59 th Interdepartmental Hurricane Conference

Advisory/Forecast Requirements • Location – Current latitude and longitude + motion – Forecast through Advisory/Forecast Requirements • Location – Current latitude and longitude + motion – Forecast through 5 days • Intensity – Current intensity to nearest 5 kt – Forecast through 5 days • Structure – Current R 34, R 50, R 64 in quadrants – Forecasts through 3 days. March 9, 2005 59 th Interdepartmental Hurricane Conference

What is available to determine TC structure (e. g. , Most of the Time) What is available to determine TC structure (e. g. , Most of the Time) • Surface observations • Satellite – IR/WV/Vis cloud drift winds • High quality, but far from the center • Spatial coverage often limited – Quikscat – surface wind vectors • Speeds good to 50 kt • Rainfall effects winds • Directions unreliable – SSM/I surface wind speeds • Rain free areas • Speed only – AMSU nonlinear balance winds • Temporal coverage limited • Not at the surface • Not available to the forecasters March 9, 2005 59 th Interdepartmental Hurricane Conference

What is Needed to determine TC Structure (sometimes in the Atlantic/E. Pacific) Aircraft + What is Needed to determine TC Structure (sometimes in the Atlantic/E. Pacific) Aircraft + Surface observations + Satellite observations = A complete 2 -d wind field + Boundary layer wind estimation model (e. g. , H*winds, Franklin et al. observations, Kepert’s model etc…) = Surface Winds March 9, 2005 59 th Interdepartmental Hurricane Conference

How to get what is needed all of the time? Answer: Satellite only 2 How to get what is needed all of the time? Answer: Satellite only 2 -d TC wind analysis (@ flight level) Needs 1. A way to estimate the winds of the TC core 2. An analysis method that can make efficient use of the existing satellite derived wind data March 9, 2005 59 th Interdepartmental Hurricane Conference

Needs 1: Estimate of Near Core TC Winds Develop relationships between Infrared (IR) satellite Needs 1: Estimate of Near Core TC Winds Develop relationships between Infrared (IR) satellite imagery and flight level winds (i. e. , Kimberly Mueller’s Poster) • IR Satellite data – Global coverage of the tropics and subtropics – Half-hourly temporal coverage (most places) • Air Force Reconnaissance Data (1995 -2003) – Measurements of winds, thermodynamics, location every 10 s – Systematic flight pattern and flight level – 87 storms, 405 independent cases March 9, 2005 59 th Interdepartmental Hurricane Conference

Hurricane FLOYD – 1515 UTC 14 Sep 99 Hurricane IRIS – 0015 UTC 9 Hurricane FLOYD – 1515 UTC 14 Sep 99 Hurricane IRIS – 0015 UTC 9 Oct 01 MSLP 932 mb MSLP 954 mb MAX Sustained Winds 125 kt MAX Sustained Winds 120 kt NE SE SW NW 64 kt 110 75 60 90 64 kt 15 15 10 15 50 kt 180 140 105 150 50 kt 25 25 15 25 34 kt 250 190 150 190 34 kt 125 50 40 60

Modified Rankine Vortex • The modified Rankine vortex is described by • Where: – Modified Rankine Vortex • The modified Rankine vortex is described by • Where: – – – V = tangential wind speed Vm = VMAX – C C = storm translational speed r = radial coordinate Rm = radius of maximum wind x is unit less, positive number derived empirically from existing infrared data March 9, 2005 Vm=100 kts, Rm=55 km, x=0. 5 59 th Interdepartmental Hurricane Conference

Inner Core Wind Analysis The Vortex Model requires: • From the Best track/advisory – Inner Core Wind Analysis The Vortex Model requires: • From the Best track/advisory – Center Position – Motion Vector – VMAX • From IR analysis – RMAX – statistically derived – X – statistically derived from Rmax, Vmax and V at 182 km March 9, 2005 59 th Interdepartmental Hurricane Conference

IR-Derived RMW March 9, 2005 V at 182 km 59 th Interdepartmental Hurricane Conference IR-Derived RMW March 9, 2005 V at 182 km 59 th Interdepartmental Hurricane Conference

Symmetric Wind Field March 9, 2005 59 th Interdepartmental Hurricane Conference Symmetric Wind Field March 9, 2005 59 th Interdepartmental Hurricane Conference

Asymmetric Wind Analysis • Storm motion is assumed to be the only contributor to Asymmetric Wind Analysis • Storm motion is assumed to be the only contributor to TC asymmetry • Use the best track mean 12 -h motion • Maximum wind is assumed to act 90 degrees to the right of motion, where March 9, 2005 59 th Interdepartmental Hurricane Conference

Needs 2: A Wind Analysis Method • Independent of the coordinate system – we Needs 2: A Wind Analysis Method • Independent of the coordinate system – we use a cylindrical framework. • Input can be weighted for reliability and density • Can utilized scalar wind or “speed only” information • Accumulates data in a storm-relative framework using the working best track and advisory forecast positions March 9, 2005 59 th Interdepartmental Hurricane Conference

Variational Wind Analysis • Find Uij Vij to minimize cost function C (Thacker and Variational Wind Analysis • Find Uij Vij to minimize cost function C (Thacker and Long 1990): C = wk[(uk-Uk)2 + (vk-Vk)2] + wm(sm-Sm)2 + [ ( r. Uij 2 + r. Vij 2) + ( Uij 2 + Vij 2] • • • Uij Vij are gridded radial and tangential wind u k, vk = obs, Uk Vk= analysis counterpart of ukvk sm, Sm are observed wind speeds and analysis counterpart Wk and Wm are data weights , terms are smoothness constraints The wind analysis is gridded function that is interpolated to observation point • azimuthal smoothing >> radial smoothing • Could also add other constraints if necessary March 9, 2005 59 th Interdepartmental Hurricane Conference

Example: Hurricane Ivan 0912 18 Z Datasets available • • • IR-based core (0 Example: Hurricane Ivan 0912 18 Z Datasets available • • • IR-based core (0 -200 km) estimate at 1800 UTC AMSU nonlinear balanced winds at 1214 UTC SSMI F 13, F 14, F 15 at 1253, 1330, 1423 UTC Quikscat at 1045 UTC Cloud drift, WV winds at 0745, 1115, 1345, 1655 UTC. March 9, 2005 59 th Interdepartmental Hurricane Conference

Datasets (continued) QSCAT AMSU ( Plot provided by Remote Sensing Systems available at www. Datasets (continued) QSCAT AMSU ( Plot provided by Remote Sensing Systems available at www. remss. com ) March 9, 2005 59 th Interdepartmental Hurricane Conference

Datasets… SSM/I (Plot provided by Remote Sensing Systems available at www. remss. com ) Datasets… SSM/I (Plot provided by Remote Sensing Systems available at www. remss. com )

Cloud drift and water vapor winds From CIMSS March 9, 2005 59 th Interdepartmental Cloud drift and water vapor winds From CIMSS March 9, 2005 59 th Interdepartmental Hurricane Conference

Combinations Combinations

Combinations (cont. ) Combinations (cont. )

Final Analysis R 34 175 180 125 185 R 50 120 115 80 125 Final Analysis R 34 175 180 125 185 R 50 120 115 80 125 R 64 80 65 60 60 R 34 150 120 100 150 R 50 100 90 70 90 R 64 80 60 45 55 From this analysis 18 Z advisory

How Does this relate to H*Wind? • Complementary (not a replacement) – Could be How Does this relate to H*Wind? • Complementary (not a replacement) – Could be a input to H*Wind • Differences – Fully automated (soon) including quality controls – Can be produced globally – Does not rely on recon – Can utilize “speed only” datasets • SSMI – Cylindrical coordinate framework March 9, 2005 59 Interdepartmental Hurricane Conference th

Future Plans In the next few months • totally automate analysis, data weighting and Future Plans In the next few months • totally automate analysis, data weighting and quality control. • Investigate options for creating surface wind estimates • Producing wind analyses every 6 hours by mid 2005 (data dependent) and display them via a web site. • Evaluate and take feedback as to its usefulness at the end of the 2005 Hurricane Season. March 9, 2005 59 th Interdepartmental Hurricane Conference