1 / 13

Walt Petersen and Kevin Knupp UAH/ESSC November 7, 2007 walt.petersen@nasa

UAH/NSSTC THOR Center and Hazardous Weather Testbed. MAX ?. KHTX. MIPS/NSSTC. 68 km. ARMOR. RSA. 75. KGWX. ARMOR. DD lobe 1 km Res. 1.5 km Res. LMA 100-500 m. MAX. MIPS Profiler. LMA Antenna. NEXRAD. KBMX.

keran
Download Presentation

Walt Petersen and Kevin Knupp UAH/ESSC November 7, 2007 walt.petersen@nasa

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. UAH/NSSTC THOR Center and Hazardous Weather Testbed MAX ? KHTX MIPS/NSSTC 68 km ARMOR RSA 75 KGWX ARMOR DD lobe 1 km Res. 1.5 km Res. LMA 100-500 m MAX MIPS Profiler LMA Antenna NEXRAD KBMX UAH THOR Center Radar Infrastructure: Exploring QPE Algorithm Development for Operational Support of TVA River Management Walt Petersen and Kevin Knupp UAH/ESSC November 7, 2007 walt.petersen@nasa.gov • Outline • Objectives for TVA-UAH interaction • Radar QPE Problem, Dual-pol solution • Brief overview of dual-pol • How do we improve QPE • UAH/NSSTC infrastructure • Simple example • data processing/flow • Rainfall algorithm • Where we are: Example products • The future • Appendix: BREAM

  2. Dual-Pol QPE Applied to Operational Hydrology • 88D Dual-pol upgrade imminent- improved QPE a primary driver. • Can dual-pol QPE replacesignificant% of TVA gauge network? • Demonstration project with ARMOR in advance of WSR-88D dual-pol upgrade • Dual-pol rain rate estimator, NO gauge input • 6-24 hour QPE over basin scales • Real time web-products • Facilitate/reintroduce radar QPE tailored to TVA needs for use in river management • Future customer specific extensions (e.g., NOAA/NWS QPE/F products, site specific terrain corrections etc. • HWT/COMET-NWS Synergies: pass products direct to WFO HUN- test utility, development E.g., Warm season precipitation event Favorable comparison to gauges- BUT much of the heaviest precip missed gauges (typical)!!! Heterogeneity of rain field presents problems for gauge-adjusted Z-R totals but not for dual-pol Walter A. Petersen, University of Alabama Huntsville

  3. Z (dBZ) vs R(mm/hr) 50 vs 100 mm/hr at 50 dBZ over a valid range of observed DSDs!! R (mm/hr) 100 0 50 Z (dBZ) QPE: Problem with conventional radar-rainfall approaches: Reflectivity Factor (Z) - Rainfall Rate (R) Relations Sample of current operational relationships: Z = 300 R1.4 - convective rain Z = 250 R1.2 - tropical rain Z = 200 R1.6 - summer stratiform rain Z = 130 R2.0 - winter stratiform (eastern US) Z = 75 R2.0 - winter stratiform (western US) • NEXRAD measures rainfall using one variable- Z – at single polarization (H) • Problem dates from 1940’s: Numerous rainfall-reflectivity relationships, which one is correct? • Errors of 100-200% are common. Why? • Measurement is sensitive to rain drop size distribution, presence of hail/ice/snow, and radar calibration. • Unacceptable errors for high resolution hydrological application (e.g., flash flood nowcasting, runoff modeling) • Even gauge corrections are still beholden to gauge calibration/error/sample mismatch- a problem at times. Proprietary information, Walter A. Petersen, University of Alabama Huntsville

  4. Polarimetric Radar Variables U.S. Research NCAR NSSL CSU NASA UND NOAA ESRL UMASS UAH ARMOR • Reflectivity factor Z at horizontal polarization • - Measure of drop size and concentration; • most sensitive to SIZE (D6) • 2. Differential reflectivity ZDR • - Measure of median drop diameter→SIZE/SHAPE • - Useful for rain / hail / snow discrimination→SIZE/SHAPE • 3. Differential phase ΦDP (Specific Differential Phase- KDP) • - Measure of content and size→NUMBER/SHAPE • - Immune to radar miscalibration, attenuation, and partial beam blockage • 4. Copolar-correlation coefficient ρhv • - Indicator of mixed precipitation → SHAPE/PHASE/CANTING (Depolarization) • - Useful for identifying non-meteorological scatterers Operational: NEXRAD, TV U.S. Broadcast Huntsville New York Houston Chicago Tampa • Advantages: Obtain a better description of particle types and shapes in a given volume of space • More accurate rain rates (improved QPE) • Hydrometeor ID and non-meteorological scatterers • Consistent calibration

  5. Dual-Pol: Improved Quantitative Precipitation Estimation (QPE) and Hydrometeor Identification (HID) Particle-Size Controlled Smaller ZDR Larger ZDR 6 mm Large Drops Small Drops 1 mm vs Insects Rain vs Smaller Z Larger Z a Hail/Graupel Rain vs b Smaller KDP Larger KDP Number Concentration Controlled Larger ZDR Smaller ZDR Smaller # Larger # Small Drops Large Drops vs Smaller KDP Larger KDP Dual-Pol Interpretation Microphysics Rain Small Drops (1 mm) Large Drops (> 4 mm) Axis ratio decreases with size- more oblate Hail/Graupel Radar “sees” Axis ratio ~ 1 • Tumbling and lower dielectric strength makes hail look like a spheres • Unless they start to melt… Melting Hail/Graupel (Toroid or ice core; looks like a huge drop) Axis ratio < 1 Proprietary information, Walter A. Petersen, University of Alabama Huntsville

  6. ARMOR: C-band Dual-Polarimetric Radar • Variables: Z, V, W, ZDR, fDP,KDP, rhv, • LDR, HID • ARMOR Standard Ops (24/7): • 3-Tilt dual-pol scan every 5 min. • 1-Tilt surveillance every 2.5 min. • Significant Weather/Research Ops: • Adaptive rapid sector/full volumes, RHI’s, vertically pointing • MIPS: Mobile Integrated Profiling System • 24/7 Ops • Vertically pointing, adaptive vertical resolution, DSD, wind, temperature, pressure, humidity profiles MAX: Mobile X-band Dual-Polarimetric Radar • Raingauges and Disdrometers • 2 Parsivel optical disdrometers, 2D Video Disdrometer (CSU) • Geonor rain gauge, three tipping bucket, 1 WXT-510 impact gauge/disdrometer, CHARM rain gauge network, TVA gauge network • MAX OPS: • Adaptive rapid sector/full volumes, RHI’s, vertically pointing • Mobile targeted QPE studies; severe wx How do we get to “improved QPE”: UAH QPE Infrastructure Walter A. Petersen, University of Alabama Huntsville

  7. rhv Z HID Cloud Drizzle Lt. Rain Mod. Rain Heavy Rain Hail Hail/Rain fDP ZDR KDP Small Hail Rain/Sm. Hail Dry Snow Wet Snow Ice Crys. How do we get to “improved QPE”? ARMOR PPI at 19:38 UTC Use of raw variables takes more work for less experienced.......... Combined polarimetric variables offer a powerful means to discriminate liquid from frozen precipitation: Improved QPE, land surface hydrology, warning decision support. Proprietary information, Walter A. Petersen, University of Alabama Huntsville

  8. How do we get to “improved QPE” UAH ARMOR Polarimetric Data Processing ARMOR T1 NSSTC/UAH Raw IRIS files (w/Vaisala HCLASS) Clean using Z, rhv, var (FDP) WFO HUN (AWIPS, GR) Level II (sweeps) UF IRIS Images Filter FDP, recompute KDP (adaptive FIR filtering) Correct Attenuation (Z) [DZ and DZDR = aKDPb] Differential Attenuation (ZDR) NetCDF Images/Tables TVA Write UF Compute Hybrid rain rates (R[KDP,ZDR,Z]) Compute HID Accumulated Rainfall Optional DD Walter A. Petersen, University of Alabama Huntsville

  9. UAH Rainfall algorithm YES R = R(ZH) GOOD DATA? KDP 0.3 and ZH  35? YES R = R(KDP) NO YES ICE PRESENT? NO ZH BAD? R = R(ZHRAIN) YES NO R=BAD R=BAD NO KDP 0.3, ZH 35.0 dBZ ZDR 0.5 dB? NO NO ZH > 30 dBZ, ZDR 0.5 dB? 1-hr Accumulation 6-hr (N-hr) Accumulation YES R = R(ZH) YES (1) R(KDP,ZDR) (2) R(KDP) (3) R(ZH,ZDR) R = R(ZH,ZDR) R > 50 mm/hr, dBZ > 50 ,or Z, ZDR corr. too large ? YES R = R(KDP) ARMOR RAIN RATE ALGORITHM YES R =R(KDP,ZDR) KDP ≥ 0.5? YES R =R(ZH,ZDR) KDP< 0.5? no no Proprietary information, Walter A. Petersen, University of Alabama Huntsville

  10. Product Access: http://www.nsstc.uah.edu/ARMOR/webimage • Operational • Current image • Last 10 image and 3-hour loops • Scan comparisons between variables • Automatic updating • Products • All dual-pol variables for first 3 sweep elevations • Hydrometeor types (fuzzy and table based) • Rainfall 1-hour (image) • Rainfall 6-hour (image) • 6-hour TVA basin rain statistics (text)

  11. UAH ARMOR RADAR 10/16/2007 (PM): 6-Hour Rainfall Accumulation Products http://www.nsstc.uah.edu/ARMOR/webimage/ • 6-Hour Rainfall Accumulation Image Product • Centered on ARMOR radar in Huntsville • TVA Basins and 25 km range rings indicated with white contours. • TVA gauge locations indicated as points • Numeric table summarizing basin mean rainfall statistics (area mean, maximum, minimum and standard deviation of 1 km pixels in each basin). • % RPxl = Percentage of basin area covered by > 0.005 inches of accumulated rainfall in the past 6-hours

  12. Future TVA Work? Tennessee River Basins (Unregulated area between Kentucky dam and mouth of Tennessee river, 710 Square Miles.) KY VA TN NC 100 km SC MS GA AL • System refinements (more products, delivery methods, verification statistics etc.) • Correction scheme for 88D (dual-pol tuning) • Test terrain following rain-map with BREAM* database (*next talk) • Topography: A problem area for radar • Leave gauges in steep terrain for now • Target work (MAX/MIPS, other leveraged opportunities) site specific 88D corrections over terrain?

  13. Future: Test Utility of High-Resolution SPC WRF BASIN SPECIFIC QPF? ROCKCASTLE AT BILLOWS 0.713 in. 0.027 in. 0.059 in. 0.367 in. 0.042 in. 0.000 in. WOLF CREEK LOCAL 0.504 in. 0.003 in. 0.069 in. 0.141 in. 0.029 in. 0.003 in. MOUTH OF OH TO BARKLEY DAM 0.000 in. 0.000 in. 0.000 in. 0.000 in. 0.000 in. 0.000 in. LAUREL R @ MUNICIPAL DAM @ CORBIN 0.255 in. 0.011 in. 0.035 in. 0.304 in. 0.068 in. 0.002 in. DOVER TO BARKLEY 0.000 in. 0.000 in. 0.003 in. 0.002 in. 0.000 in. 0.000 in. MOUTH OF TN TO KY DAM 0.000 in. 0.000 in. 0.000 in. 0.007 in. 0.000 in. 0.000 in. CUMB AB WILLIAMSBURG 0.074 in. 0.071 in. 0.090 in. 0.306 in. 0.111 in. 0.089 in. NORTH FORK HOLSTON RIVER NEAR GATE CITY 0.003 in. 0.873 in. 0.000 in. 0.774 in. 0.557 in. 0.603 in. CADIZ TO MOUTH 0.000 in. 0.000 in. 0.000 in. 0.000 in. 0.000 in. 0.000 in. CLINCH RIVER ABOVE TAZEWELL 0.021 in. 0.801 in. 0.026 in. 0.422 in. 0.331 in. 0.527 in. . . . . . . . . . . . . . .

More Related