TAVE OUTLINE

TAMDAR AERIBAGO Validation Experiment (TAVE) - Memphis

Wayne Feltz, Erik Olson, John Short, Sarah Bedka, Kristopher Bedka, Tim Wagner, and Scott Cultice

University of Wisconsin - Madison SSEC/CIMSS

TAVE OUTLINE

• TAVE Experiment Description

• TAVE Dataset Overview

• Preliminary Validation Results

• Future Analysis and Deployment

• Preliminary RUC vs Aircraft Validation

TAVE LOGISTICS

Location: Memphis airport at Air National Guard

Dates/Timing: 23 February - 08 March 2005

Weather: Dry, cool, two periods of rainfall, very little in the way of high dewpoint events/convection

Data: Made available in near real-time, netcdf format from anonymous ftp server

Radiosondes - 49 total launches

12 Tail Numbers with TAMDAR instrumentation (excluding the 3 with bad sensors)

Web site: http://cimss.ssec.wisc.edu/tamdar/

Memphis Airport Location

Memphis ANG DeploymentMemphis ANG Deployment

Memphis ANG DeploymentMemphis ANG Deployment

AERIBAGO INSTRUMENTATION

• Vaisala DigiCORA III RS-92 GPS Sounding System

• Atmospheric Emitted Radiance Interferometer (AERI)

• GPS Receiver

• VAISALA Surface PTU Station

• VAISALA 25K Ceilometer

• http://cimss.ssec.wisc.edu/aeribago/

Instrument Summary

• Radiosondes - Vertical temperature and water vapor, Vaisala RS-92 radiosondes, 20 preparation time, 1 1/2 hour flight time, realtime monitoring capability

• Surface Station - Vaisala sensors, one minute, p, t, q, v, u, tested and calibrated for accuracy at SSEC

• AERI - Vertical temperature and water vapor profiles to three kilometers at ten minute resolution within PBL

• GPS - Integrated total water vapor at half hourly time resolution must be at a stationary location for at least ten days

• Vaisala Ceilometer - Cloud base heights every ten seconds

TAVE AERIBAGO SETUP

Vaisala DigiCORA-III ReceiverRS-92SGP Capable

Radiosonde Profile vs TAMDAR

GPS Receiver

Vaisala Surface Station

Vaisala Ceilometer

Atmospheric Emitted RadianceInterferometer (AERI)

AERI Thermodynamic Retrievals

Other Datasets Collected

• TAMDAR/ACARS Profiles from MADIS and converted to netcdf over Memphis

• RUC Analysis Profiles over Memphis

• Satellite, radar, and surface imagery

• GOES pwv, profiles

All Available at:

ftp://ftp.ssec.wisc.edu/validation/exper/tamdar/data/

PreliminaryPreliminaryTAMDAR TAMDAR Thermodynamic Thermodynamic ValidationValidation

TAVE Validation Datasets

Radiosonde Schedule

Saab 340s arrive / depart in the following groups (local time):

7:35 - 7:55 Arrive 9:20 - 9:55 Depart

12:36 - 1:15 Arrive 14:15 - 15:00 Depart

17:25 - 18:00 Arrive 19:10 - 19:50 Depart

TAVE Radiosonde Launch Time Slots:

8:30, 10:15, 14:15, 17:00, and 19:00 Local time

Up to 5 Launches per day

Careful attention was paid to keep the radiosonde out of sun and allowing sensor to acclimate to outside environment for several minutes before launch

Radiosonde Launch Procedure

Radiosonde Ascent Rate

5,000 ft, reached 5 minutes after launch. 1000 ft/min ascent rate.

10,000 ft, reached 12 minutes after launch. 800 ft/min ascent rate.

24,000 ft, reached 30 minutes after launch. 700 ft/min ascent rate.

30,000 ft, reached 43 minutes after launch. 500 ft/min ascent rate.

These balloons are smaller and lighter than the ones the NWS launches at 00Z and 12Z at numerous sites across the US.

Profiles Comparison Examples

Profiles Comparison Examples

Profiles Comparison Examples

Excluded Tail Numbers

Tail numbers: 5580, 5598 - bad relative humidity information

5552 - bad temperature information

Radiosonde vs TAMDAR Trajectories

Dashed - DescendingSolid - AscendingBlack lines - State Boundaries

Profile Examples

Statistics for Ascent/Descent

Temperature

Statistics for Ascent/Descent

Relative Humidity

Statistics for All Matches

Future Validation Improvements

• Validation quality of radiosonde water vapor measurements by correlating with Bago surface sensor and AERI radiance intercomparison

• TAMDAR sensor QC should be looked at with more scrutiny, maybe rh1 or rh2 should be used instead of rh once consensus between TAMDAR sensor indicates good rh1 or rh2 data

• Validation matches should be screened in the vertical as a function of horizontal distance between sensor to help account for horizontal variability of the water vapor

Future Validation ImprovementsContinued

• Temperature from ACARS measurements within +/- 30 minutes should be included on rms/bias plots

• Time criteria can be tightened with more matches from future deployment

• Wind validation using GPS winds from radiosondes vs TAMDAR/ACARS measurements

TAVE Summary

• Preliminary Validation is underway

• TAVE Data Available: http://cimss.ssec.wisc.edu/tamdar

• More extensive validation will be presented at Virginia meeting in September

• Next Deployment possible in May or July/August in Memphis

Comparison of TAMDAR and non-TAMDAR with RUC 1 hr forecast

Source statistical data provided by Bill Moninger – FSL

http://acweb.fsl.noaa.gov/ruc_acars/

Data from weekly ‘fit’ statistics fromMid-January - Mid-March 2005

Ralph PetersenCIMSS

Comparison of TAMDAR and non-TAMDAR with RUC 1 hr forecast

Bias Temperature (oC) Standard Deviation

Source statistical data provided by Bill Moninger - FSL

WMO Temperature Requirements for Regional NWP (RMS) - 0.5 oC, Minimum Acceptable 3.0 oC

Data from Mid-January - Mid-March 2005

Comparison of TAMDAR and non-TAMDAR with RUC 1 hr forecast

Bias Speed (m s-1) Standard Deviation

Source statistical data provided by Bill Moninger - FSL

WMO Wind Requirements for Regional NWP (RMS) - 1 m s-1, Minimum Acceptable 5 m s-1

Data from Mid-January - Mid-March 2005

Comparison of TAMDAR and non-TAMDAR with RUC 1 hr forecast

Bias Relative Humidity (%) Standard Deviation

Source statistical data provided by Bill Moninger - FSL

WMO Moisture Requirements for Regional NWP (RMS) - 5%, Minimum Acceptable 20%

Data from Mid-January - Mid-March 2005

Summary of Comparison of TAMDAR and non-TAMDAR with RUC 1 hr forecastEvaluations performed between TAMDAR and all other aircraft

- Need to separate reports from similar parts of atmosphere- Need to view in combination with Rawinsonde Co-Locations- Need to have more Q/C information

- WMO criteria used here to exclude outliers- May have been too conservative

Two data sets show notably different characteristics- Temperature

- More TAMDAR outliers (> WMO Specs) anddifferent (cold) bias

- Wind- More TAMDAR outliers and

Bias ( 3x ) and Standard Deviation larger- Moisture

- No WVSS-II comparison made yet- Bias increases as RH increases- RMS near WMO limits (careful since we are comparing to

model forecast/analysis)

Issue – Who will manage, maintain and distribute TAMDAR Q/C info?- Currently, WMO requires NWS to do this for other aircraft data.