Basic Principles of Weather Radar

Basic Principles of Weather Radar

Dr. Scott M. Rochette

Basis of Presentation

• Introduction to Radar• Basic Operating Principles• Reflectivity Products• Doppler Principles• Velocity Products• Non-Meteorological Targets• Summary

(http://www.weathermatrix.net/radar/education/articles/laughlin/images/KDFX.jpg)

Laughlin AFB, TX (KDFX)0612 UTC 26 May 2001

Radar

• RAdio Detection And Ranging• Developed during WWII for detecting enemy

aircraft• Active remote sensor

– Transmits and receives pulses of E-M radiation– Satellite is passive sensor (receives only)

• Numerous applications– Detection/analysis of meteorological phenomena– Defense– Law Enforcement– Baseball

Weather Surveillance Radar

• Transmits very short pulses of radiation – Pencil beam (narrow cone) expands outward– Pulse duration ~ 1 μs (7 seconds per hour)– High transmitted power (~1 megawatt)

• ‘Listens’ for returned energy (‘echoes’)– Listening time ~ 1 ms (59:53 per hour)– Very weak returns (~10-10 watt)

• Transmitted energy is scattered by objects on ground and in atmosphere– Precipitation, terrain, buildings, insects, birds, etc.– Fraction of this scattered energy goes back to radar

(http://www.crh.noaa.gov/mkx/radar/part1/slide2.html)

(http://www.crh.noaa.gov/mkx/radar/part1/slide3.html)

(University of Illinois WW2010 Project)

(University of Illinois WW2010 Project)

http://weather.noaa.gov/radar/radinfo/radinfo.html

Determining Target Location

• Three pieces of information– Azimuth angle– Elevation angle– Distance to target

• From these data radar can determine exact target location

Azimuth Angle

• Angle of ‘beam’ with respect to north

(University of Illinois WW2010 Project)

Elevation Angle

• Angle of ‘beam’ with respect to ground(University of Illinois WW2010 Project)

Distance to Target

• D = cT/2• T pulse’s round trip time

(University of Illinois WW2010 Project)

Scanning Strategies 1

• Plan Position Indicator (PPI)– Antenna rotates through 360° sweep at

constant elevation angle– Allows detection/intensity determination of

precipitation within given radius from radar– Most commonly seen by general public– WSR-88D performs PPI scans over several

elevation angles to produce 3D representation of local atmosphere

Plan Position Indicator

• Constant elevation angle• Azimuth angle varies (antenna rotates)

(University of Illinois WW2010 Project)

Elevation Angle Considerations

• Radar usually aimed above horizon– minimizes ground clutter– not perfect

• Beam gains altitude as it travels away from radar

• Radar cannot ‘see’ directly overhead– ‘cone of silence’– appears as ring of minimal/non-returns around

radar, esp. with widespread precipitation

• Sample volume increases as beam travels away from radar

• Red numbers are elevation angles• Note how beam (generally)

expands with increasing distance from radar

(http://weather.noaa.gov/radar/radinfo/radinfo.html)

• Blue numbers are heights of beam AGL at given ranges

• Most effective range: 124 nm

Scanning Strategies 2

• Range Height Indicator (RHI)– Azimuth angle constant– Elevation angle varies (horizon to near

zenith)– Cross-sectional view of structure of specific

storm(University of Illinois WW2010 Project)

Radar Equation for Distributed Targets 1

P

c P g l K

r

N D

vr

A

t

B C

i ii

k

D

3 2

2

2

2

6

1

1 0 2 4 2ln

where Pr average returned power A numerical constants B radar characteristics C target scatter efficiency characteristics D equivalent radar reflectivity factor (Ze)

Choice of Wavelength 1

Pr 1

2– Typical weather radar range: 0.8-

10.0 cm– WSR-88D: ~10 cm– TV radar: ~5 cm

Choice of Wavelength 2

Pr 1

2– Pr inversely proportional to square of wavelength (i.e., short

wavelength high returned power)– However, shorter wavelength energy subject to greater attenuation

(i.e., weaker return signal)– Short wavelength radar better for detecting smaller targets

(cloud/drizzle droplets)– Long wavelength radar better for convective precipitation (larger

hydrometeors)

Radar Equation for Distributed Targets 2

where Pr average returned power

Rc radar constant

Ze equivalent radar reflectivity factor (‘reflectivity’)

r distance from radar to target

PR Z

rrc e 2

Radar Equation for Distributed Targets 3

• Pr is: - directly proportional to ‘reflectivity’- inversely proportional to square of distance between

radar and target(s)

PR Z

rrc e 2

Equivalent Radar Reflectivity Factor 1

where Ni number of scattering targets

Di diameter of scattering targets v pulse volume

v

DNZ

k

iii

e

1

6

Equivalent Radar Reflectivity Factor 2

• Ze relates rainfall intensity to average returned power

• ‘Equivalent’ acknowledges presence of numerous scattering targets of varying:– sizes/shapes– compositions (water/ice/mixture)– distributions

• Several assumptions made (not all realistic)

Equivalent Radar Reflectivity Factor 3

• Ze is: - directly proportional to number of scatterers- inversely proportional to sample volume- directly proportional to scatterer diameter raised to 6th power

- Doubling size yields 64 times the return

v

DNZ

k

iii

e

1

6

(University of Illinois WW2010 Project)

dBZ

• Typical units used to express reflectivity• Range:

• –30 dBZ for fog • +75 dBZ for very large hail

dBZZ

mm me 1 0

11 0 6 3lo g

Scanning Modes• Clear-Air Mode

– slower antenna rotation– five elevation scans in 10 minutes– sensitive to smaller scatterers (dust, particulates, bugs, etc.)– good for snow detection

• Precipitation Mode– faster antenna rotation– 9-14 elevation scans in 5-6 minutes– less sensitive than clear-air mode– good for precipitation detection/intensity determination

– trees

• Livestock– birds– bats– insects

• Other– sun strobes– anomalous propagation

Clear-Air Mode

Precipitation Mode

Greer, SC (KGSP)(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

Clear-Air Mode Precipitation Mode

Reflectivity Products 1

• Base Reflectivity– single elevation angle scan (5-14 available)– useful for precipitation detection/intensity

• Usually select lowest elevation angle for this purpose

– high reflectivities heavy rainfall• usually associated with thunderstorms• strong updrafts larger raindrops• large raindrops have higher terminal velocities• rain falls faster out of cloud higher rainfall rates• hail contamination possible > 50 dBZ

Reflectivity Products 2

• Composite Reflectivity– shows highest reflectivity over all

elevation scans– good for severe thunderstorms

• strong updrafts keep precipitation suspended

• drops must grow large enough to overcome updraft

Little Rock, AR (KLZK)Precipitation Mode

Base Reflectivity Composite Reflectivity

Z-R Relationships 1

Z aR b

where Z ‘reflectivity’ (mm6 m-3) R rainfall rate (mm h-1) a and b are empirically derived

constants

Z-R Relationships 2

• Allow one to estimate rainfall rate from reflectivity

• Numerous values for a and b– determined experimentally– dependent on:

• Precipitation character (stratiform vs. convective)• Location (geographic, maritime vs. continental,

etc.)• Time of year (cold-season vs. warm season)

Z-R Relationships 3Relationship Optimum for: Also recommended

for:

Marshall-Palmer(Z=200R1.6)

General stratiform precipitation

East-Cool Stratiform(Z=130R2.0)

Winter stratiform precipitation - east of

continental divide

Orographic rain - East

West-Cool Stratiform(Z=75R2.0)

Winter stratiform precipitation - west of

continental divide

Orographic rain - West

WSR-88D Convective(Z=300R1.4)

Summer deep convection

Other non-tropical convection

Rosenfeld Tropical(Z=250R1.2)

Tropical convective systems

(WSR-88D Operational Support Facility)

(WSR-88D Operational Support Facility)

Z-R Relationships 4

Reflectivity

Marshall-Palmer

(Z=200R1.6)

East-Cool Stratiform(Z=130R2.0)

West-CoolStratiform(Z=75R2.0)

WSR-88D Convective(Z=300R1.4

)

Rosenfeld Tropical

(Z=250R1.2)

15 dBZ 0.25 mm h-1 0.51 mm h-1 0.76 mm h-1 <0.25 mm h-1

<0.25 mm h-1

20 dBZ 0.76 mm h-1 1.02 mm h-1 1.27 mm h-1 0.51 mm h-1 0.51 mm h-1

25 dBZ 1.27 mm h-1 1.52 mm h-1 2.03 mm h-1 1.02 mm h-1 1.27 mm h-1

30 dBZ 2.79 mm h-1 2.79 mm h-1 3.56 mm h-1 2.29 mm h-1 3.30 mm h-1

35 dBZ 5.59 mm h-1 4.83 mm h-1 6.60 mm h-1 5.33 mm h-1 8.38 mm h-1

40 dBZ 11.43 mm h-

1

8.89 mm h-1 11.68 mm h-

1

12.19 mm h-1

21.59 mm h-1

45 dBZ 23.62 mm h-

1

15.49 mm h-

1

20.58 mm h-

1

27.94 mm h-1

56.39 mm h-1

50 dBZ 48.51 mm h-

1

27.69 mm h-

1

36.58 mm h-

1

63.50 mm h-1

147.32 mm h-

1

55 dBZ 99.82 mm h-

1

49.28 mm h-

1

65.02 mm h-

1

144.27 mm h-1

384.56 mm h-

1

60 dBZ 204.98 mm h-1

87.63 mm h-

1

115.57 mm h-1

328.42 mm h-1

1004.06 mm h-1

Radar Precipitation Estimation 1

• 1-/3-h Total Precipitation– covers 1- or 3-h period ending at time

of image– can help to track storms when viewed

as a loop– highlights areas for potential (flash)

flooding– interval too short for some applications

Radar Precipitation Estimation 2

• Storm Total Precipitation– cumulative precipitation estimate at time of image– begins when radar switches from clear-air to

precipitation mode– ends when radar switches back to clear-air mode– can help to track storms when viewed as a loop– helpful in estimating soil saturation/runoff– post-storm analysis highlights areas of R+/hail– no control over estimation period

St. Louis, MO (KLSX)

1-h Total Precipitation(ending at 2009 UTC 11 June 2003)

Storm Total Precipitation(0708 10 June 2003 to 2009 UTC 11 June 2003)

Radar Precipitation Estimation Caveats

• No control over STP estimation interval• Based on empirically-derived formula

– not always ideal for given area/season/character

• Hail contamination– (large) water-covered ice pellets very reflective– causes overestimate of precip intensity/amount

• Mixed precipitation character in same area– convective and stratiform precipitation falling

simultaneously– which Z-R relationship applies?

• Patterns generally good, magnitudes less so

Doppler Effect

• Based on frequency changes associated with moving objects

• E-M energy scattered by hydrometeors moving toward/away from radar cause frequency change

• Frequency of return signal compared to transmitted signal frequency radial velocity

(http://www.howstuffworks.com/radar1.htm)

(Williams 1992)

(http://www.crh.noaa.gov/mkx/radar/part1/slide13.html)

Radial Velocity 1

• Hydrometeors moving toward/away from radar– Positive values targets moving away from

radar– Negative values targets moving toward radar

• Can be used to ascertain large-scale and small-scale flows/phenomena– fronts and other boundaries– mesoscale circulations– microbursts

Radial Velocity 2

• Base Velocity– ground-relative– good for large-scale flow and straight-

line winds• Storm-Relative Velocity

– storm motion subtracted from radial velocity

– good for detecting circulations and divergent/convergent flows

Houston, TX (KHGX)warm colors away from radar

cool colors toward radar

Base Velocity Storm-Relative Velocity

(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

Buffalo, NY (KBUF)1944 UTC 28 April 2002Storm-Relative Velocity

mesocyclone

(http://www.srh.weather.gov/jetstream/remote/srm.htm)

The Doppler Dilemma 1

• Pulse can only travel so far and return in time before next pulse is transmitted– Distant targets may be reported as close,

and/or– Velocities may be aliased

• Pulse Repetition Frequency (PRF)– transmission interval– typical values 700-3000 Hz (cycles s-1)– key to determining maximum unambiguous

range (Rmax) and velocity (Vmax)

The Doppler Dilemma 2

• Maximum Unambiguous Range (Rmax)– Longest distance between target and

radar that can be ‘measured’ with confidence

– Inversely proportional to PRF

PRF

cR

2max

The Doppler Dilemma 3

• Maximum Unambiguous Velocity (Vmax)– Highest radial velocity that can be

‘measured’ with confidence– Directly proportional to radar

wavelength and PRF

4max

PRFV

The Doppler Dilemma 4

4max

PRFV

PRF

cR

2max

8maxmax

cRV

The Doppler Dilemma 5

• If Ractual > Rmax, range folding occurs– distant echoes appear close to radar– Rapp = Rmax – Ractual

– second-trip echoes

• If Vactual > Vmax, velocity folding occurs– radial velocities misreported– Vapp = - (2Vmax – Vactual)– Sign of Vmax = Vactual

The Doppler Dilemma 6

• If Vmax = ± 25 m s-1 and target is moving away from radar at 30 m s-1 – i.e., Vactual = +30 m s-1

• Vapp = - (50 – 30) = -20 m s-1 – toward radar at slower speed!

• What about target moving at - 50 m s-1?• Vapp = - [-50 – (- 50)] = 0 m s-1

– Target would appear to be stationary!

• If Rmax is large, then Vmax has to be small (and vice versa)– cannot be large simultaneously!

Non-Meteorological Targets

• Ground Clutter– trees– mountains– buildings

• Livestock– insects– birds– bats

• Other ‘Targets’

Ground Clutter

• Stationary objects usually filtered out

• Swaying trees or towers may show up

• Look for drifting high reflectivity returns near radar

Cannon AFB, NM (KCVS)Precipitation Mode

(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

Mountain Blockage

• Low elevation angle scans blocked by terrain

• ‘Shadows’ appear consistently in imagery

• Mainly a problem in western U.S.

Boise Mountains

Owyhee Mountains

Boise, ID (KBOI) Clear-Air Mode

(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

WSR-88D Network

Building Blockage

• Nearby building blocks beam if building is taller than antenna (~100 ft)

• Narrow ‘shadows’ appear consistently in imagery

• Occurs in/near metropolitan areas

Houston, TX (KHGX)Precipitation Mode

(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

Livestock

• Radar can be used to track migrations

• Insects tend to be ‘carried’ by low-level flow

• Bats and birds can travel ‘against’ flow

Greer, SC (KGSP)Clear-Air Mode

Insects flying on NE winds(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

Blue: Low Bird DensityGreen: High Bird Density

Fort Polk, LA (KPOE)Precipitation Mode

(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

Brownsville, TX (KBRO)Clear-Air ModeRaptor Migrations (S N)(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

Laughlin, TX (KDLF)Precipitation Mode

Bat Roost Rings(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

Other Targets 1

• Sun strobes– occur typically around dawn/dusk– radar receives intense dose of E-M

radiation along narrow radials– similar strobes occur if beam

intercepts intense source of microwave radiation• other radars• microwave repeaters

National Radar MosaicPrecipitation Mode

Sun Strobes(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

Other Targets 2

• Anomalous propagation (AP)– beam refracted into ground under very

stable atmospheric conditions • inversions• near large bodies of water• behind thunderstorms

– appear similar to intense precipitation• compare to surface observations• check satellite imagery• examine higher elevation scans

Melbourne, FL (KMLB)Clear-Air Mode

Anomalous Propagation (AP)(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

(http://www.crh.noaa.gov/mkx/radar/part2/slide31.html)

Other Targets 3

• Chaff– metallic dust deployed by aircraft– used to diffuse radar signatures– often seen in vicinity of military bases

• Aircraft– large metallic objects– excellent scatterers– original intended radar target

Melbourne, FL (KMLB)Clear-Air Mode

Chaff(http://virtual.clemson.edu/groups/birdrad/COMMENT.HTM)

Shreveport, LA (KSHV)Clear-Air Mode

0954 CST 1 February 2003

Summary

• Weather surveillance radar has varied uses– short-term weather forecasting– hazardous weather warnings– hydrologic applications– tracking bird migration patterns

• Must be aware of radar’s limitations– WYSINAWYG– What You See Is NOT ALWAYS What You Get!

(University of Illinois WW2010 Project)

(http://www.aero.und.edu/~rinehart/cartoons.html)