Presented by Sandra Cruz-Pol,

1
CASA an NSF ERC

• Presented by Sandra Cruz-Pol,
• Professor Electrical and Computer Engineering
• UPRM CASA PI
• Aug 9, 2006 ONR Visit to UPRM

2
There is insufficient knowledge about what is
actually happening (or is likely to happen) at
theEarths surface where people live. NRC 1998
3
CASA dense networks of low power radars
Colorado State University Commonwealth of
Massachusetts IBM Mount Holyoke College National
Science Foundation NOAA/National Weather
Service Oklahoma Climatological
Survey OneNet Raytheon Company Rice
University Texas Medical Center University of
Delaware University of Massachusetts University
of Oklahoma University of Puerto Rico University
of Virginia Viasala Vieux and Associates

• Year 3 of a 10 year program
• Initial 5 year investment 42 M
• (includes 17M Engineering Research Center grant
  from NSF)
• 6-7M per year annual cash budget
• Critical site visit review April 2006 in MA

4
Todays Radar Networks
Comprehensive Coverage gt 3 km
5
Sensors Required for US Nation-Wide Coverage
NetRad -OTG
300 m floor
3 km floor
6
Projects IP1, IP2, IP3
Rain mapping, distributed hydro. modeling, flood
predicting response in urban zone.
IP1
Wind mapping (100s m resolution, 10s second
update) for detecting, pinpointing, forecasting
wind events 30 km node spacing.
IP2
Rain, Urban Flooding (Houston)
Wind, storm prediction (Oklahoma)
IP3
Rain, mountainous terrain (Puerto Rico student
led)
Off-the-Grid Radar Network for QPE over complex
terrain, student-led project
7
Project IP1 - Initial 4-Node Test Bed

• Annual storm climatology for 7,000 sq km test bed
  region
• 4 tornado warnings (2 touchdowns)
• 50 thunderstorms

8
User Driven System Design

• Users NWS Forecast Office, Emergency Managers,
  atmospheric scientists will use the Oklahoma test
  bed
• Severe weather severe thunder storms, hail, and
  tornados impacts 90 of EMs in Oklahoma.
• Tornado Pinpointing cited by EMs as important for
  managing deployment and protection of first
  responders.
• Tornado Anticipation cited by NWS and EMs as most
  important for increasing lead time.
• All users cited more frequent updates of radar
  data as a critical need.
• There is a need for lower troposphere, high
  resolution data for detecting convergence lines,
  gust fronts, straight line winds.

Sources Structured surveys (N72) of Oklahoma
Emergency Managers In-Depth Interviews (N37) of
EMs and NWS using snowball sampling and content
analysis to extract information test bed user
group.
9
NEXRAD
gt 3 km covered by current technology
3 km
NEXRAD Map winds, rain above 3 km (10,000)
10
NetRad System underneath NEXRAD
3 km
25 km
NetRad IP1 Goal Map winds below 3 km with 500 m
resolution
Water spout at Mayaguez, PR- Sept 2005
11
NetRad Elevation coverage
7
6
6 km
5
4
3 km
3
2
1
25 km
Goal IP1 - Map winds below 3 km. 2o pencil
beam antenna yields median 500 m resolution 7
elevation beam positions scan 0-14 degrees
12
Cone of silence observed by neighboring radar
7
6
5
4
3 km
3
2
1
25 km
Goal Map winds below 3 km. 7 elevation beam
positions scan 0o-14o Neighbor radars map cone
of silence above a radar. Multiple-Doppler wind
measurement throughout.
13
NetRad Sampling Modes
Limited sector Mode
Sit-and-Spin Mode
Samples the Atmosphere When, Where the End-User
Need is Greatest
14
NetRad adaptive data pull
End users weather services, emergency response
15
Prototype IP1 Radar
11x14x23 in.
Elevation Scan Ball-screw linear actuator Range
- 5o to 30o Scan 20o/sec
Azimuth Scan Mfr Kollmorgan Scan
120o/sec Acceleration
16
Tour of the IP1 Sites

• Avg. Separation 25.3 km
• Coverage 6947 km2
• 98 coverage below NEXRAD
• 41 coverage is dual-Doppler (2850 km2)
• 25 coverage below 250 m
• Avg. AGL NetRad 364 m
• Avg. AGL NEXRAD 1000 m

17
Where are we now?

• IP1 Project End-to-End DCAS network of 4 rapid
  scan radar nodes.
• 2 pol magnetron Radars cost 200k in parts
  replacement cost insurance coverage was 1.5 M
  for 4 radars during shipment to OK.
• Custom towers tower-top positions to host
  radars.
• Infrastructure
• Weight 1,500
• Site tower top
• HVAC, radome
• Ethernet, fiber, 802.11 access to node
• Software closed-loop, MCC, policy mechanism
  but no decision-based policy as yet.

Est. 500k to buy install these radars
18
IP3 Student Led Test Bed in Puerto Rico The
Off-the-Grid Network
19
IP3 Student Led TestBed in Puerto Rico The
Off-the-Grid Network
2-D video disdrometer deployed at SJ NWS and at
UPRM to characterize rain statistics during
normal rain and T.S. Jeanne and Frances R-Z
relation cal
20
Puerto Rico Testbed IP3

• Update1st radar is here http//casa.ece.uprm.edu

Recent interest from Argentina
21
Weather Research and Tracking (WeatherRats K12
Initiative)
22
CASAs Vision
Revolutionize our ability to observe, understand,
predict and respond to weather hazards by
creating DCAS networks that  sample the
atmosphere where and when end-user needs are
greatest.
touching peoples lives... saving
lives/property, reducing vulnerability, providing
economic benefits through improved warning and
response to hazards diverse education,
outreach industrial opportunities, commercial
development
23
Plans for next 5 years
Goal System build-out beyond 4 nodes.
OK System Test-Bed
IP5 - 2nd Gen. NetRad System
Technology Goals escan panel radars bistatic,
Fabry
MA Technology Test-Bed (and PR Tech Test-Bed)
IP4 - CLEAR
PR Technology Test-Bed
Goal QPE in irregular terrain minimal
infrastructure system energy balance education
IP3 OTG/Complex Terrain
FR Technology Test-Bed
Goal QPE, closed DCAS loop via hydro models
reduce infra. costs
IP2 Rain Urban Flooding
OK System Test-Bed
Goal 1st end-to-end system use rapid mscan to
quantify value of DCAS (ie, extra 10 dB)
IP1 Wind and storm prediction
1/09
1/10
1/11
1/12
1/13
1/04
1/03
1/05
1/06
1/07
1/08
24
We are open to collaboration
Mi CASA es tu CASA
25
Contacts

• Dr. Sandra Cruz-Pol- Microwave Remote Sensing and
  atmospheric attenuation
• Dr. José Colom Microwave Radars Circuits
• Dr. Rafael Rodríguez Microwave Antennas
• Dr. Wilson Rivera- Wireless networks
• Dr. Walter Díaz Social Sciences
• Dr. Mario Ierkic Atmospheric phenomena
• Dr. Héctor Monroy EM propagation
• Dr. Lionel Orama Power
• All emails webs are on http//ece.uprm.edu

26
Very Low Cost Phased Array Radars - Semiconductor
Cost
Si wafer mask design setup
80k
30k
100 Radar Buildout Semiconductor Cost 8M GaAs
vs. 3M Si