new weather-surveillance capabilities for nssl’s … weather-surveillance capabilities for...
TRANSCRIPT
New weather-surveillance capabilities for NSSL’s phased-array radar
Sebastián Torres,Ric Adams, Chris Curtis, Eddie Forren, Igor Ivić, David Priegnitz, Eddie Forren, Igor Ivić, David Priegnitz, John Thompson, and David Warde
CIMMS/The University of Oklahoma and
National Severe Storms Laboratory/NOAA
Norman, OK (USA)
ERAD 2012Toulouse, France
28 June, 2012
More
Frequent Low
Elevation
Scans
Elevations
Below 0.5 deg
10%
Other
8%
No New Scans
4%
what do users want?
62%of forecasters* want
faster updates
Scans
37%
Faster Scans
25%
Better Range
Unfolding
16%
*US National Weather Service survey on radar scanning strategy needs (2008)2
fundamental trade-offs
For example, if we want faster updates, we can’t have both
better spatial coverage and better precision.3
Update Time
mission: MPAR
the NWRT PAR is being exploited to demonstrate
improved weather surveillance capabilities
upgrades
hardware
software
dsp
demonstrations
fast adaptive scanning
multi-function
capabilities
improved data
quality
4
focused observations faster updates
storm 2 storm 3
el
conventionalscan
adaptivescan
�
from www.gapolitico.com
storm 1
az900
it’s all about cutting the “waste”!
PAR
5
Adaptive DSP Algorithm for PAR Timely Scans
how does ADAPTS work?
active beam positions meet one or more criteria
• Elevation angle
• Significance
• NeighborhoodReal-time display of active beam positions
azaz
el
el
ADAPTS classifies beam positions as active or inactive
1st Criterion1st Criterion 2nd Criterion2nd Criterion 3rd Criterion3rd Criterion
• Neighborhood
Zth
continuity and coverage
Always scan at the lower elevations
Range
Z
Determine significant weather signals
Real-time display of active beam positions
Look around to respond to storm
evolution
EL
AZleft right
up
down
6
ADAPTS cuts the “waste”!
AZ
EL
Radar scans full volume
tn
AZ
EL
tn+1
AZ
EL
tn+2
AZ
EL
AZ
EL
Radar scans active beam positions
tn+3
AZ
EL
AZ
EL
Radar scans full volume
ADAPTSdetermines active beam positions
ADAPTSdetermines active beam positions
ADAPTSredetermines active beam positions
ADAPTSredetermines active beam positions
ADAPTSresets active
beam positions
ADAPTSresets active
beam positions
ADAPTSdetermines active beam positions
ADAPTSdetermines active beam positions
full volume
time
FULL ADAPTS ADAPTS ADAPTS ADAPTS FULL ADAPTS
Surveillance update time
Scanning strategy schedule: periodic surveillance
tn tn+1 tn+2 tn+3
7
active beam positionsfull volume
ADAPTS “keeps an eye” on the sky
AZ
EL
ADAPTSdetermines active beam positions
ADAPTSdetermines active beam positions
ADAPTSredetermines active beam positions
ADAPTSredetermines active beam positions
tn
AZ
EL
tn+1
AZ
EL
tn+2
AZ
EL
tn+3
AZ
EL
tn+4
AZ
EL
tn+5
AZ
EL
Radar scans active beam positions with full dwells and
Radar scans full volume
time
FULL ADAPTS ADAPTS ADAPTS ADAPTS ADAPTS
Scanning strategy schedule: continuous surveillance
tn tn+1 tn+2 tn+3
8
ADAPTS
tn+4 tn+5
active beam positions with full dwells and inactive beam positions with short dwells
full volume
periodic vs. continuous surveillance
Periodic Full Scans
�inactive beam positions not continuously scanned
• need periodic surveillance
• a full scan runs every ~5 min
• delayed detection of developing
Continuous Surveillance
�inactive beam positions continuously scannedwith shorter dwell times
• achieve continuous surveillance
• data used for detection only
• timely detection of developing
9
• delayed detection of developing storms
• beam positions at low-elevation angles forced to be active
• timely detection of developing storms
• beam positions at low-elevation angles may be active
ADAPTS reduces update times NWRT PAR – 05 Jun 2008 – 19:51 to 20:45 UTC
10
a line of storms develops along a cold front over central OK
time savings depend on size and location of storms
Tem
po
ral R
eso
luti
on
(m
in)
1.1
1.2
1.3
1.4
a) 1 May 2009 Nominal Update Time
Tim
e Savings
0104:35 0129:24 0154:27 0217:27 Time (UTC)
Tem
po
ral R
eso
luti
on
(m
in)
0.9
1.0
0040:59
0040:59 0129:24 0217:27b)
200 km 150 km 100 km 200 km 150 km 100 km 200 km 150 km 100 km
ADAPTS Update Times
11
theoretical performance
Conventional Scan
conventional
update tim
e (m
in)
12
periodic surv. (5 min)
periodic surv. (10 min)
periodic surv. (20 min)
continuous surv.
ADAPTS with continuous surveillance provides the fastest updates and the most timely detection of new storms.
ave
rage update tim
e
storm coverage (% of 90-deg sector, all elevations)
summary
• users want faster updates
– focused observations result in faster updates with no loss in data quality or spatial coverage
• phased-array radars are suited to perform adaptive focused observations
– not constrained by mechanical inertia– not constrained by mechanical inertia
– ADAPTS demonstrates that PARs can achieve performance levels not feasible with current operational technology
for more information:http://cimms.ou.edu/~torres/par.html
13