Global Atmospheric Turbulence Decision Support System for Aviation

Global Atmospheric Turbulence Decision Support System for Aviation John Williams, Bob Sharman, and Cathy Kessinger (NCAR); Tony Wimmers and Wayne Feltz (UW/CIMSS) NASA Applied Sciences Weather Program Review Boulder, CO November 18, 2008 and UW-Madison SSEC/CIMSS

Global Turbulence project goals • Develop global probabilistic turbulence nowcasts and forecasts and supporting global convective nowcasts – Address clear-air turbulence (CAT), convectively-induced turbulence (CIT), and mountain-wave turbulence (MWT) – 0 -3 hour nowcasts of convection – 0 -36 hour forecasts of turbulence, 10, 000 – 45, 000 ft. • Support World Area Forecast System (WAFS) – Provide improved accuracy and timeliness over current SIGMETs and SIGWX charts • Support Next. Gen SWIM 4 -D data cube • Leverage and provide outlet for previous FAA and NASA-funded work (e. g. , Oceanic Weather)

Approach • Develop a database of global turbulence reports for empirical model development, tuning and verification • Adapt CONUS GTG (AWRP-funded) diagnostics for use with Global Forecast System (GFS) model data • Adapt Oceanic Nowcast (AWRP/NASA) for global application on GOES East and West, MTSAT, and Meteosat • Adapt satellite-based turbulence algorithms (NASA/GOES-R) including tropopause folding, CIT and MWT features • Adapt CIT diagnostics (AWRP/NASA) for use with GFS model and satellite convection diagnoses and nowcasts • Develop nowcast/forecast data fusion system • Verify and benchmark the forecasts vs. SIGMETs • Run real-time demonstration over GOES East/West domain – Adapt CONUS Experimental ADDS display (AWRP/NOAA) – Cockpit uplink messages (prev. AWRP-funded) – Web-based pilot/forecaster feedback

Flow Chart for Global GTG Geo-, Leo- Satellites Convection Diag. & Now. GFS Model CIT/CAT Diagnoses In Situ Observations (tuning) Turb. Detection (trop. folding) Global Graphical Turbulence Guidance Mountain Wave

Global GTG Demonstration at NCAR Global GTG ATM / Dispatchers Synthetic SIGMETs Pilots World Wide Web

Future NEXTGen Use of Global GTG ATM / Dispatchers Pilots WAFC NEXTGen 4 -D Data Cube Human over the Loop SIGMETs World Wide Web

Cockpit Uplink Demo • Aircraft-relative display of cloud top height • Pilot and dispatcher receive a “heads up” for approaching weather (common situational awareness) ASCII display via cockpit printer Future Positions • United Airlines test on oceanic flights Cloud Top Height >40 kft Flight Path 30 -39 kft Current Position

Empirical Turbulence Data: UAL EDR United EDR above 10, 000 ft MSL, 07 -01 -2008 to 07 -15 -2008

Empirical Turbulence Data: Delta EDR above 10, 000 ft MSL, 07 -01 -2008 to 10 -31 -2008

Empirical Turbulence Data: Ude, various airlines, 11 -1 -2008 to 11 -5 -2008

Empirical Turbulence Data: AIREPs

Turbulence data collection status • Most data are available from NCAR archives – NCDC archive data may also be used • Quality control development is underway • Database design is underway

Adapting GTG diagnostics for GFS • Challenges of global model RUC (hybrid. B coordinates) – Cycle boundaries in longitude – Poles are singular points – Constant lat-lon grid, therefore Δx (long. ) nonuniform in spatial distance, much smaller near poles – Different vertical coordinates – Some diagnostics break down at equator GFS (sigma coordinates)

GFS-GTG 6 -hr forecasts valid 18 Z 4 Nov 2008 (election day) Ellrod index FL 350 EDR index FL 350 Ri from thermal wind FL 200 Note: breaks down at equator GFS-based diagnostic RUC-based diagnostic

GFS-GTG status • Initial implementation of common GTG core software (supports RUC, WRF, and GFS models) • Still testing GFS-based diagnostics – Need to refine based on turbulence reports for particular cases – Then perform statistical evaluations over extended time period • GFS data are being ingested and converted in real-time – 0. 5 degree GFS 00 -, 03 -, 06 -, 09 -, and 12 -hour pressure coordinate files (will switch to 1/3 degree) • Real-time system monitoring tools have been set up

(Near-) Global Convection Nowcasting MTSAT-1 R Courtesy of David Johnson • Geostationary satellite-based methodologies cover +70 degrees latitude • Could expand coverage using polar-orbiting satellites – Northern latitudes benefit (over-the-pole aviation routes) – Few (no? ) commercial aviation routes over Antarctica

Geo-Satellite Methodology Cloud Top Height Diagnosis Interest Global Convective Diagnosis Oceanic • Fuzzy logic data fusion of two methods – Cloud Top Height from longwave infrared (10 micron) & GFS model – Global Convective Diagnosis [TB(6. 5 micron) – TB(10 micron)] • Convective Diagnosis Oceanic (CDO) Interest Field (0 to 2) – Final, binary product will have a threshold applied

Challenges for Convection Nowcasting • Varying, full-disk satellite scanning strategies – 3 hourly: GOES-E and GOES-W – 1 hourly: MTSAT-1 R – 15 minutes: Meteosat-9 • GOES satellites have partial scans that could be merged – Discontinuities across scanning boundaries make extrapolation difficult • Planning for 3 -hourly updates for Global Turbulence – Provides an analysis and 3 -hr nowcast • Will examine Meteosat-9 imagery in anticipation of future rapid-update GOES-R capability

Global convective nowcasting status • Real-time ingest and processing of GOES full disc scans (Terrascan) implemented • Initial (uncalibrated) CDO field produced in real-time • Archived data from MTSAT has been transmitted from CIMSS to NCAR for analysis • Real-time system monitoring tools have been set up

Tropopause Fold Turbulence Prediction (Tony Wimmers, UW/CIMSS) • Background • Algorithm basics • Adapting the algorithm to aircraft hazard awareness • Latest developments 20

Background

Background: Tropopause folding and Clear Air Turbulence (CAT) Upper-air front stratosphere 12 200 300 subtropical air mass 10 tropopause 400 8 6 500 600 700 14 fro nt polar air mass (~100 km) From Shapiro, M. A. (1980): Turbulent mixing within tropopause folds as a mechanism for the exchange of chemical constituents between the stratosphere and the troposphere, J. Atmos. Sci. , 37, 994 -1004. 4 Height (km) Pressure (h. Pa) 150

Background: GOES Layer-Average Specific Humidity (GLASH) (WV channel) (GLASH product) Wimmers, A. J. , and J. L. Moody, A fixed-layer estimation of upper tropospheric specific humidity from the GOES water vapor channel: Parameterization and validation of the altered brightness temperature product, Journal of Geophysical Research-Atmospheres 106 (D 15),

Background: Synoptic-scale UTH

Background: Synoptic-scale UTH 300 h. Pa Potential Vorticity

Background: Validation with aircraft ozone lidar Tropopause folds were measured at crossings of upper-troposphere air mass boundaries Wimmers, A. J. , J. L. Moody, E. V. Browell, J. W. Hair, W. B. Grant, C. F. Butler, M. A. Fenn, C. C. Schmidt, J. Li, and B. A. Ridley, Signatures of tropopause folding in satellite imagery, Journal of Geophysical Research - Atmospheres, 109, art. no. 8360, 2003.

Algorithm basics

latitude decreasing specific humidity 1. Humidity product (cloud masked) longitude

latitude decreasing specific humidity 2. Smoothed gradient of humidity product longitude

latitude decreasing specific humidity 3. Contour along boundaries longitude

decreasing specific humidity latitude 4. Extend polygons out from boundaries Wimmers, A. J. and J. L. Moody, Tropopause folding at satellite-observed spatial gradients: 2. Development of an longitude empirical model, Journal of Geophysical Research – Atmospheres, 109, art. no. D 19307, 2004.

Adapting the algorithm to aircraft hazard awareness

Adapting to aircraft hazard awareness § Validation data - Eddy Diffusion Rate § Automated reporting of inertial disturbance on commercial aircraft § 3 -minute integration time per measurement (short). This indicates several measurements through any single turbulent region. § Collected and quality-checked by NCAR § ~ 1 month latency 33

x x longitude decreasing specific humidity latitude Adapting to aircraft hazard awareness 1. Filter for the trop. folds associated with turbulence 2. Assign a vertical height to the trop. folds

Filtering cases of tropopause folding Measurements of turbulence are limited to near-orthogonal crossings of the aircraft with the direction of flow “ 20º” “ 75º” Also limiting factors: - Direction of flow - WV gradient strength - Time of year (possibly)

Potential temperature difference from the lower tropopause Height assignment (validation with EDR data) Predicted area of turbulence Distance from tropopause break probability of turbulence EDR reports conform to the expected cross-section of turbulence in a tropopause fold

Recent developments

Algorithm Development Adaptation for operational use § In 2007, the Tropopause Folding Turbulence Product was adapted as an algorithm for the GOES Algorithm Working Group (AWG) § AWG is a multi-year project to prepare end-user products to be optimized and online at the very start of the GOESR operations § The transition of this algorithm to the GOES-R platform includes the following: § Port original Matlab code to Fortran ( ) § Incorporate products into GEOCAT environment ( ) § Apply to synthetic GOES-R 6. 19 um channel data § Validation of the adapted product with EDR data 38

Algorithm Development Adaptation for operational use Test case output: Tropopause Fold minimum/maximum heights

Algorithm Development Adaptation for operational use Test case output: Tropopause Fold exposed flight directions

Algorithm Development Future objectives – Adapt the algorithm for use with MTSAT and Meteosat imagery § Makes the product truly global § Gradients will be intercalibrated with MODIS data and tropopause features can be confirmed with the Aqua Ozone Mapping Instrument (OMI) 41

Summary (John Williams) • The Global Turbulence DSS project aims to produce global probabilistic turbulence nowcasts and forecasts supported by global convective nowcasts – Improve WAFS SIGMET and SIGWX charts – Provide turbulence information for pilots and airtraffic management via the Next. Gen 4 -D data cube • Leverages existing CONUS systems and NASA- and FAA-funded turbulence and convection R&D • Progress made on empirical turbulence database, GFS -GTG, global convection diagnosis, and tropopause fold detection. • An example of a comprehensive, end-to-end, collaborative product development project • Next. Gen transition to operations needs to be defined

Integrated System Solution (ISS) Diagram