Transcript Global Weather Services in 2025

Acquisition and Use of NEXRAD
and FAA Doppler Weather Radar
Data
Presented to the Monroney Aeronautical Center
9 November 2000
ns f
Kelvin K. Droegemeier
Center for Analysis and Prediction of Storms
and
School of Meteorology
University of Oklahoma
NEXRAD Doppler Radar Network
NEXRAD Facts and Figures

158 radars (141 in the Continental US)
– 120 National Weather Service radars
– 26 Department of Defense radars
– 12 Federal Aviation Administration radars
NEXRAD Data Types




Archive Level I (raw receiver data)
Level II data (digital data in spherical
coordinates at full resolution)
Archive Level III (digital products)
Archive Level IV (forecaster-generated products)
NEXRAD Data Types




Archive Level I (raw receiver data)
Level II data (digital data in spherical
coordinates at full resolution)
Archive Level III (digital products)
Archive Level IV (forecaster-generated products)
NEXRAD Product Data (NIDS)



24 products available from all CONUS radars
in real time
Lowest 4 elevation angles only
Low-precision because values are quantized
(e.g., 0-5, 5-10, 10-15)
ns f
ns f
NEXRAD Data Types




Archive Level I (raw receiver data)
Level II data (digital data in spherical
coordinates at full resolution)
Archive Level III (digital products)
Archive Level IV (forecaster-generated products)
NEXRAD Base (Level II) Data

Full resolution digital data
– Full data precision
– All elevation angles
Not available in real time except for
selected sites (more on that later)
 This data set is the focus of our
efforts

ns f
Base Data Usage: NSSL Warning
Decision Support System on 3 May 1999
ns f
Courtesy National Severe Storms Laboratory
Trimmed Detections and Ground Truth Damage Paths
Hits (142)
Misses (25)
FAs (21)
Courtesy D. Zittel
The Value of NEXRAD Radar Data for
Numerical Storm Prediction:
The 3 May 1999 Oklahoma
Tornado Outbreak
Copyright 1999 The Daily Oklahoman
CAPS Numerical Forecasts of the May 3 Tornadic Storms
5:00 pm - Model Initialization Time
Storm Beyond Velocity
Range of NEXRAD
ARPS Prediction Model
(0 hour forecast)
NEXRAD Radar Observations
ns f
CAPS Numerical Forecasts of the May 3 Tornadic Storms
5:30 pm - 30 min Forecast
Model Generates
the Storm Itself
ARPS Prediction Model
(1/2 hour forecast)
NEXRAD Radar Observations
ns f
CAPS Numerical Forecasts of the May 3 Tornadic Storms
6:00 pm - 1 hour Forecast
ARPS Prediction Model
(1 hour forecast)
NEXRAD Radar Observations
ns f
CAPS Numerical Forecasts of the May 3 Tornadic Storms
6:30 pm - 1.5 hour Forecast
Strong Mesocyclone
Present
ARPS Prediction Model
(1 1/2 hour forecast)
Tornado on the
Ground
NEXRAD Radar Observations
ns f
CAPS Numerical Forecasts of the May 3 Tornadic Storms
7:00 pm - 2 hour Forecast
ARPS Prediction Model
(2 hour forecast)
NEXRAD Radar Observations
ns f
Forecasts With and Without NEXRAD Data
WITHOUT
WITH
Moore, OK
Tornadic
Storm
2-Hour CAPS Computer Forecast
Down to the Scale of Counties
NEXRAD Radar Observations
Summary: WSR-88D Radar Data


The scientific and operational communities need
base data (real time and archived)
Although NIDS data are available in real time from
all WSR-88D radars, they are insufficient for many
applications (NWP, hydrology)
– Degradation of precision
– Only the lowest 4 tilts are transmitted

Base data currently are not available in real time
– Originally would have been expensive
– Presumed large volume of data (10 mbytes/5 min/radar)
– Need wasn’t there 10 years ago

The technology and need now exist to prototype
the direct acquisition, use, and archival of base
data in real time
ns f
The Collaborative Radar
Acquisition Field Test (CRAFT)

Establish a prototype real time WSR-88D base data
acquisition test bed to
– Evaluate strategies for compressing and transmitting
base data in real time
– Develop efficient and cost-effective strategies for direct
digital ingest, archive, and retrieval at NCDC
– Assess the value of base data in numerical weather
prediction
– Test web-based data mining techniques for rapid
perusal/access of base data by the scientific community
ns f
Technical Strategy
At the radar site
Dedicated 56K line
Cisco 1600
Series
Router($2000)
($2000 - $6000/year)
Server
WSR-88D
Internet
RIDDS
Linux PC
Unidata LDM
($1500)
Users
Repeater Hub
ns f
Original CRAFT Network
ns f
CRAFT Phase I: Proof-of-Concept
ns f
Abilene Network
January 1999
Seattle
New York
Sacramento
Denver
Indianapolis
Kansas City
Los Angeles
Atlanta
Abilene Router Node
Abilene Access Node
Operational January 1999
Planned 1999
Houston
New Concept: Abilene/Internet2 +
NEXRAD
ns f
New Concept: Abilene/Internet2 +
NEXRAD
ns f
New Concept: Abilene/Internet2 +
NEXRAD
ns f
New Concept: Abilene/Internet2 +
NEXRAD
U-WA
NCEP
AWC
NCAR/FSL
NCDC
OU
TPC
ns f
Radars Now Delivering Data
ns f
Links to be Established by This Time Next Year
ns f
Using the Data for Aviation
ns f
Weather Hazard Detection: FAA Earmark
Funding to CAPS
ns f
Some Examples
GOES Visible, 2245 Z
4 June 1998
KFWS Composite Reflectivity
00 Z, 4 June 1998
ns f
Sample Aviation Products
Cloud Type and LWC
at FL 050
Cloud Type and LWC
at FL 320
Cloud Type and LWC
N/S X-Section
ns f
Sample Aviation Products
Downburst Potential
Surface Isotachs &
Streamlines
CAPE & Helicity
ns f
Sample Aviation Products
Surface Visibility
Clear-Air Turbulence
Icing Potential
ns f
Weather Hazard Detection: FAA Earmark
Funding to CAPS
ns f
Weather Hazard Detection: FAA Earmark
Funding to CAPS
ns f
Statistical Climatologies of Storm Characteristics
(location, intensity, movement, initiation, decay)
relative to NAS assets
Pugh (2000)
Mitchell et al. (2000)
Mitchell et al. (2000)
The Future: FAA Radars

The CRAFT concept can be extended to
include FAA radars that process weather
information
– TDWR (terminal Doppler weather radar)
– ASR (airport surveillance radar)
– ARSR (air route surveillance radar)
ns f
TDWR
ns f
Weber (2000)
Airport Surveillance Radars (ASR-9)
ns f
Weber (2000)
Airport Surveillance Radars (ASR-9)
ns f
Weber (2000)
Airport Surveillance Radars (ASR-11)
ns f
Weber (2000)
Air Route Surveillance Radars (ARSR-4)
ns f
Weber (2000)
The Result

An integrated, national data set of highly
detailed weather radar information for use in
– Numerical weather prediction
– Real time air traffic control and planning
– Research of specific relevance to aviation


The radar data can be used to create
“assimilated” data sets that provide all
meteorological variables at high resolution
We’re positioning Norman to serve as a
national data repository for real time access
ns f
Funding Status
– NOAA ESDIM Grant funded (CAPS+NSSL+OSF+NCDC)


$540K/3 years
Research Thrusts
– Test of direct ingest/archival at NCDC
– Improve compression algorithms
– Initial work on web-based data mining
– NOAA earmark funding to OU




$474K for 1 year
Expand CRAFT to 30 radars (CRAFT-2)
Develop data assimilation capabilities for the WRF model
Kelvin doing a mini-sabbatical at the NSSL this fall
– HPCC Proposal ($150K for 1 year, about to be funded)



Data mining
Network quality of service research
Hardware for additional radars
ns f
Funding Status
– FAA earmark funding to OU






$250K for 1 year
Assimilate data from multiple radars
Provide real-time aviation hazard products
A collaboration with the SPC and AWC
Hope to involve NCAR (say via the National Convective
Forecast Product)
Fits into the CCFP?
ns f
The Next Steps







The Oklahoma meteorological community is ideally
poised to take the lead in bringing the FAA radars into
Project CRAFT
Will involve collaboration with others (MIT/LL, NCAR)
The need has been recognized (Weber, 2000)
A proof-of-concept test is needed (cost will be
minimal)
Could begin with OKC TDWR and extend to other
systems (one ASR-9, one ASR-11, one ARSR-4)
The FAA earmark grant could be used to add a TDWR
to the CRAFT data in the real time assimilated data
sets
Develop a white paper and establish collaborations
ns f