Real-Time Radar Processing at the National Severe Storms

Real-Time Radar Processing at the National Severe Storms Laboratory Kurt D. Hondl NSSL Radar Research & Development Division

NMQ & WDSS-II • What capabilities have been developed at NSSL that could support NOAA’s Next. Gen effort? An integrated, real time, multi sensor platform to develop, test, and assess advanced techniques in quality control, data integration, application development (in severe weather and precipitation estimation), display, and short term forecasting A 3 D/4 D grid of radar (and other observations) and derived severe weather & precipitation estimation products The result of 10+ years of research, application development, and operational testing at NSSL & NWS Forecast Offices Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 2

NMQ: National mosaic and Multi sensor QPE http: //nmq. ou. ed • A suite of quantitative precipitation estimation algorithms • Uses multi-sensor observations • Products delivered to RFCs (AHIPS) and NWS FOs (FFMPA) for operational use • A 2 -yr online archive a precipitation products Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 3

NMQ Precipitation Products • Reflectivity Products Vertical Profile of Reflectivity, Bright band Identification Hybrid Scan Reflectivity (VPR corrected) • Precipitation Products 1 HR Precip, 3 HR, 6 HR, 12 HR, 24 HR, and 72 HR Precip Radar only, Multi Sensor, Radar with Gauge Bias Correction Local Gauge Bias Gauge only Precip • Precipitation Verification Comparison to Stage II, Stage IV, HADS, MPE References Vasiloff, S. , D. J. , Seo, K. Howard, J. Zhang, D. H. Kitzmiller, M. G. Mullusky, W. F. Krajewski, E. A. Brandes, R. M. Rabin, D. S. Berkowitz, H. E. Brooks, J. A. Mc. Ginley, R. J. Kuligowski, and B. G. Brown. , 2007: Q 2: Next generation QPE and very short term QPF. Bull. Amer. Met. Soc. 88, 1899 1911. Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 4

WDSS-II: Warning Decision Support System – Integrated Information http: //wdssii. nssl. noa • A suite of data ingest and quality control processes • Severe weather algorithms and applications • Uses multi-sensor observations • Products delivered to NCEP (SPC and AWC) and NWS FOs (OUN, FTW, TUL), CAPS, ESRL, NCAR and Environment Canada for operational use • An archive of automated Hail Swath and Rotation Track information Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 5

WDSS-II Severe Weather Algorithms • 3 D Reflectivity Mosaic Automated QC Neural Network VIL, Echo. Top, Composite, Isotherms Hail (Max Expected Size, Probability, Hail Swath) Kmeans/Segmotion Nowcast, Storm Tracking Storm Classification • 3 D Velocity Vortex Detection & Diagnosis Algorithm (from Azimuthal Shear) Multi Doppler Wind Analysis • Multi Sensor Algorithms Lightning (NLDN & LMA) Near Storm Environment (from RUC numerical model analyses) Satellite Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 6

NSSL 3 D/4 D Radar System 3 D Radar Merger product Ingesting data from national network of WSR 88 D radars Grid resolution of 1 km x 1 km over the CONUS (6200 km x 3000 km x 20 km) with 31 vertical levels Data streaming in from multiple radars and go through an automated QC to remove non precipitation echoes Integrate with other meteorological data to generate algorithm products (satellite, RUC model analysis, etc) 3 D grid and products updated every 5 minutes References – – – Lakshmanan, V. , T. Smith, K. Hondl, G. Stumpf, A. Witt, 2006: A Real Time, Three Dimensional, Rapidly Updating, Heterogeneous Radar Merger Technique for Reflectivity, Velocity and Derived Products. Weather and Forecasting, 21, 802 823. Langston, C. , J. Zhang, and K. Howard, 2007: Four Dimensional Dynamic Radar Mosaic. J. Atmos. Oceanic Technol. , 24, 776 790. Zhang, J. , K. Howard, and J. J. Gourley, 2005: Constructing three dimensional multiple radar reflectivity mosaics: examples of convective storms and stratiform rain echoes. J. Atmos. Oceanic Technol. , 22, 30 42. Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 7

WSR-88 D Radar Mosaic and Products Generating 1 km CONUS products every 5 min Requires ~22 servers to produce 1 km CONUS 3 D radar observation grid every 5 min Another ~10 servers to produce WDSS II severe weather products and ingest multi sensor data Another ~16 servers to produce NMQ precipitation estimate products Expanding Linux server farm to add capacity Will eventually produce 0. 5 km CONUS products every 2 min to take advantage of WSR 88 D Super Resolution Will be adding Dual Polarization algorithms/products as the WSR 88 D network is upgraded Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 8

Other Radars Canadian Radars TV Radars-KPIX CASA-KLWE TDWR-OKC Adding other radars to the 3 D/4 D radar observations grid will help fill in coverage gaps not sampled by the WSR 88 D or augment observations when maintenance or outages prevents WSR 88 D observations. Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 9

NMQ/WDSS-II Domain ~150 WSR 88 D 31 Canadian Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 2 TDWR 1 TV station radar 10

NMQ/WDSS-II Collaborators and Product Distribution NMQ/WDSSII Aviation, Severe weather Precipitation Products And Diagnostics Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 11

NMQ/WDSS-II Computational Infrastructure The entire NMQ/WDSS-II processing system is composed of 2 Quad-Core AMD Opteron Processors (2. 9 GHz with 32 GB RAM) Linux servers from a single manufacturer (HP); Servers and server drives can quickly be configured, maintained and swapped out; Running Red. Hat 64 -bit OS; Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 12

Radar Data Ingest WSR-88 D • Ingest base level 2 data from 158 tri-agency (DOC, DOD, DOT) 10 cm WSR-88 D radars • Ingest base level data from 33 5 cm radars from Environmental Canada/NCDC • Ingest base level data from two TDWRs using direct connections • Ingest commercial radars in addition to experimental radar systems such as CASA and PAR Canadian Radars • System designed to accommodate increasing access to radar systems and radar networks TDWR Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 13

NMQ/WDSS-II Mosaic and Product Creation Radar Ingest WSR-88 D 4 D grids & products Precipitation Products Aviation Next. Gen Weather Products Severe Weather Products Canadian TDWR Multi-sensor Ingest Products are disseminated in Net. CDF, binary, AWIPS, N-AWIPS, GIS, and HRAP formats using the LDM protocol. Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 14

NMQ/WDSS-II Multi-Sensor Inputs Ingest and process other sources of information to improve products Radar data (as previously described) Satellite NWP data (RUC model) Lightning (NLDN, LMA) Surface Observations (ASOS, Mesonets) Rain Gauge Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 15

NMQ/WDSS-II Next. Gen Product Generation Current to 2012 Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 2012 to 2016 16

NMQ/WDSS-II Next. Gen Product Generation Current to 2012 Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 2012 to 2016 17

NMQ/WDSS-II Next. Gen Product Generation Current to 2012 Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 2012 to 2016 18

4 D Radar Mosaic Development Pathway FY 10: NCEP Implementation of CONUS 4 D reflectivity mosaic and products – – – Severe weather and aviation related products per Next. Gen requirements Resolution of 1. 0 km x 2 min update cycle with 3 D reflectivity mosaic 31 levels Ingest includes commercial, Canadian radars and TDWRS for comparison • – Evaluation of data quality concerns and attenuation at shorter wavelengths Q 2 precipitation type and rate products FY 11: Operational 0. 5 -km CONUS 4 D reflectivity mosaics and products – – – Improved automated quality control methods for TDWR, Canadian and commercial radars Initial 3 D Mosaics for Hawaii / Alaska regions Evaluating and advancing Dual Pol algorithms specific to Next. Gen requirements FY 12: Dual Polarization 4 D CONUS Mosaics prototype at NSSL – – Initial 4 D Dual Polarization Hydrometeor CONUS mosaic for model data assimilation Advanced Dual polarizing quality control and hydrometer classification FY 13: Operational Dual Polarization 4 D CONUS Mosaics at NCEP – Initial connections to Mexican & Caribbean radars Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 19

4 D Radar Mosaic Development Pathway FY 14: North America 4 D Mosaics prototype at NSSL – – Inclusion of additional gap filling radars in 4 D Mosaic (Commercial, CASA) Mexican & Caribbean radars in 4 D Mosaic FY 15: Operational North America 4 D Mosaics at NCEP – – Integrate airborne radars (e. g. Hurricane observations) into the 4 D radar cube Developing strategies for 4 D radar mosaic of MPAR radar data Initial radar 3 D mosaics from European radar networks Initial radar 3 D mosaics from Asian radar networks FY 16: Hemispheric 4 D reflectivity Mosaics prototype at NSSL – Prototype the utilization of airborne radars as node functions to autonomously exchange radar with surrounding aircraft and ground systems as well as into NCEP’s 4 D dynamic mosaic FY 18: Hemispheric 4 D Radar Mosaics at NCEP – – Initial Northern Hemispheric radar mosaics Initial hemispheric precipitation and storm type mosaic combining satellite and radar FY 20: Hemispheric 4 D Radar Mosaics and Multi-sensor Aviation products at NCEP Prototype combined surface, airborne, space radars observations Initial inclusion of MPAR for 4 D mosaics Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 20

NMQ/WDSSII Next. Gen Future Research Activities Seamless Integration of Radar Advances, Systems, and Networks 2008 2009 2010 2011 2012 Canada Network 2013 2014 201 5 201 6 201 7 201 8 201 9 202 0 202 1 202 2 Commercial Radars WSR 88 D High Res TDWR Network WSR 88 D Do. D L 2 Mexico Network WSR 88 D Dual Polarization Caribbean Network CASA MPAR 1 km/5 min 1 km/2 min 500 m/1 min 250 m/1 min North America Resolution and Refresh Rate Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 21

NMQ/WDSS-II Summary • NSSL continues to facilitate the seamless, scientifically sound, high resolution integration of radar systems and networks into a 4 D frame as the basis for Next. Gen radar authority • The experimental system serves as the foundation for other aviation research and product development • Completed the initial operational infusion at NCEP for model data assimilation and CONUS, high resolution multi sensor radar based products. • In FY 10, NSSL will establish at NCEP a stand alone Next. Gen radar 4 D cube as the foundation to facilitate radar derived products to meet FAA Next. Gen requirements. Next. Gen System of Systems Workshop (Hondl / NSSL) 6 -7 22