Radar at the BNF

Basics of Radar & ARM Radar Applications with an Eye Towards the Bankhead National Forest (BNF) Site

Scott Giangrande

Meteorologist

19 May 2025

ARM Summer School 2025 at BNF

Radar Applications in Cloud-Process Studies

• There are basic unanswered questions about the physics of clouds, the interaction between clouds and incoming solar radiation and outgoing terrestrial radiation.

​

• There is still a lack of detailed cloud observations. Insufficient detail to unravel the physical processes occurring in clouds.

​

• ARM deployed new tools (including radar) for observing clouds emerged in the early 1990’s to complement satellite efforts.

Cloud radar at the ARM Southern Great Plains (SGP), Lamont, Oklahoma, USA.

The World’s Most Comprehensive Radar Research Network

• ARM operates 26 radars at its 6 ground-based facilities. These radar provide profiles, & in the case of the 11 scanning radars, 3D distributions of hydrometeors.

​

• To detect particles spanning a broad range in size and to study both microphysics and spatial morphology, the radar have varied characteristics:
• Emitted pulses with wavelengths ranging from 3 mm to 33 cm
• Zenith-pointing & scanning
• Single & dual polarization

ENA SACR image, courtesy of Brad Isom

Radio Detection And Ranging

• Detecting via radio echoes, including determining media direction, range and additional characteristics.
• Scatterers include hydrometeors (drizzle, rain, hail, snow), insects, birds.
• Dates back to ~1900: first recorded demonstrations of river boat detection (Germany, England).
• Common radar applications for atmospheric sciences include:
• Precipitation estimation and echo classification
• Cloud boundaries and fractional coverage
• Severe storm identification, tracking
• Dynamics, air motions in clouds
• Turbulence, wind sheer detection

ENA SACR image courtesy of Brad Isom

Conventional and Polarimetric Radars

The radar is collecting insights (received voltage) from an ensemble of hydrometeors over a discrete period of time.

​

Returns are influenced by propagation effects and backscattering.

​

Pulsed coherent radars emit EM radiation (short pulses at a high rate, e.g., 1000s of pulses per second).

Radar: Basic Terminology

• Transmission: A transmitter generates electromagnetic signals (e.g., pulsed) that are radiated into space by an antenna.

​

• Scattering/Doppler: A portion of the transmitted energy is intercepted by the media, radiated (scattered) in many directions:
• Rayleigh Scattering when D << λ
• Mie Scattering when D ~ λ
• Doppler effect: There is a frequency shift of the echo signals due to the Doppler effect. This frequency shift is proportional to the velocity of the target relative to the radar.

​

• Receiving: The radiation scattered back to the radar is collected by the antenna, which delivers it to a receiver.

​

• Signal Processing: The received signals are then processed to extract desired information (echo power, radial velocity, etc.).

ARM Radar

• KAZR (Vertically pointing, 35 GHz):
• Profiling cloud properties. Insights into cloud particle distributions.
• SACR (Scanning Cloud Radar, 10/35/94 GHz):
• First of a kind technology; Macroscale cloud field properties, structure.
• XSAPR/CSAPR (Scanning Precipitation Radar, 3/5-cm wavelength):
• Macroscale structure, cell tracking

Radar Code

Frequency Wavelength Radio Band Designation

300-3000 kHz (1 km-100 m) MF (medium frequency)

3-30 MHz (100-10 m) HF (high frequency)

30-300 MHz (10-1 m) VHF (very high frequency)

300-3000 MHz (1 m-10 cm) UHF (ultra high frequency)

3-30 GHz (10-1 cm) SHF (super high frequency)

30-300 GHz (1 cm-1 mm) EHF (extremely high frequency)

Frequency Wavelength IEEE Radar Band Designation

1-2 GHz (30-15 cm) L Band (original search radars, “Long”)

2-4 GHz (15-7.5 cm) S Band (“Short”, albeit only at that time)

4-8 GHz (7.5-3.75 cm) C Band (“C is for Compromise”)

8-12 GHz (3.75-2.50 cm) X Band (“… marks the spot”)

12-18 GHz (2.5-1.67 cm) Ku Band (‘K, under’ absorption band)

18-27 GHz (1.67-1.11 cm) K Band (Kurtz, ‘short’ in German)

27-40 GHz (1.11 cm-7.5 mm) Ka Band (‘K, above’ absorption band)

Radar Typically Adopt Similar Scanning Patterns

The Most Common Scanning Radar Displays

Plan Position Indicator - PPI

Range Height Indicator - RHI

• Surveillance
• Macroscale organization
• Low-level precipitation mapping

• Vertical structure, evolution
• Detailed (high resolution) core / updraft plume properties

How to Interpret Radar Measurements: A Practical Example For Radar Rainfall Estimation

Want: Cloud properties, distributions, dynamics, etc., to improve process representation.

​

​

Modeled raindrop size distributions (DSD) often assumed to follow simple forms:

N(D) = N0 exp(-λD) (exponential)

​

​

Rain properties monitored by radar primarily using the Radar Reflectivity Factor, or Z

​

​

(*some assumptions later) The estimated Z is related to the rain DSD (as its 6^th moment):

Z ~ ∫ N(D) D⁶ dD

How to Interpret Radar Measurements: A Practical Example For Radar Rainfall Estimation (Continued…)

• Rainfall Rate R may also be expressed in terms of the rain DSD [kg m^-2 s^-1] as:
• R = ∫ m(D) N(D) v(D) dD (raindrop mass)*(DSD)*(drop fall speed)
• ∫ (π/6)*ρ*D³* N0 exp(-λD) * cD^f dD

​

• Z = ∫ N(D) D⁶ dD can also again be simplified as:
• ∫ x^(v-1) * exp(-mX) dx = 𝚪(v)/m^v (gamma function)
• Z = N0 𝚪(7)/λ⁷
• R = (π/6)*ρ*N0*c * 𝚪(4+f)/λ(4+f)

​

• Substitute for λ:
• Z ~ C*R^7/(4+f)
• Z ~ C*R^1.5

​

• NEXRAD R(Z): Z = 300R^1.4

Atmospheric Studies With Radar Still Reflect Several Compromises

“Cloud”, “Weather”, “Precipitation” Radar?

• “Cloud” radar: Sensitive to cloud-sized droplets; Attenuation is a concern.

​

• “Weather” radar: Able to penetrate deeper convective cores; Less sensitive to cloud-sized media.

​

• UHF/VHF Profilers: Bragg scatter (turbulent fluctuations of the index of refraction caused by temperature and moisture fluctuations) allows for other retrieval possibilities.

ARM Radars and ARM Deep Convective Campaigns

• 2006: TWP ICE (Mather, May, Jacob)
• 2008: CASA/ARM Partnerships (Wiscombe)
• 2010-2011: MC3E (Jensen et al.)
• 2011: AMIE/DYNAMO (Long et al.)
• 2014/5: GoAmazon2014/5 (Martin et al.)
• 2018-2019: CACTI (Varble et al.)
• 2022-2023: TRACER (Jensen et al.)

Prior to BNF, the primary long-term ARM deep convective observations were those collected by the Southern Great Plains (SGP) Oklahoma facility, in the heart of “tornado alley”.

​

In addition to SGP, ARM has supported several deep convective field campaigns:

Beyond the “Soda Straw”: Pushing for 3D Observations

Atmospheric Radiation Measurement (ARM)

“Field of Beams”

Models Struggle to Capturing Rainfall:

Uncertainty In Microphysics

Fan et al. (2017) example from the BNL-led MC3E Campaign

• Representing rainfall is a challenge for finer-scale models.

​

• Most models overestimate the intensity of “convective” rainfall.

​

• This is even though weather models have many advantages towards capturing these details.

Vertical Velocity In Deep Convection: The Midlatitude Continental Convective Clouds Experiment (MC3E)

Collocation of multi-Doppler radar with profiling radar retrievals.

​

ER-2 aircraft profiling of Thunderstorms.

Jensen et al. 2016; North et al. (2017)

Models, Radar and Velocity Observations: 20 May 2011

• Microphysics “reasonable” for capturing precipitation.

​

• Results highly-coupled with the model tuning process.

​

• All simulations overestimated updraft strength.

​

• Modelers know their models are wrong… Focus on how to improve those models.

Fan et al. (2017)

Immediate returns from MC3E when adding vertical velocity as a new constraint.

Median Updraft

BNF Deployment To The Southeast U.S.

• 5+ year deployment, with operations beginning October 2024
• Abundant, locally-forced shallow to deep convective clouds
• Strong local coupling of land surface with atmospheric processes

​

New Questions Emerging: What is the Nature of Updrafts?

RWP-Estimated Vertical Velocity

Broader updraft structure consisting of successive rising “bubbles”

Observations of moist updrafts suggest that they are made up of “bubbles.”

Recent LES efforts also support thermal-type behaviors.

Varble et al. 2014

Confronting Storms with New Observational Techniques

IR Satellite

Weather Radar

Optimized Interrogation of Storms

E. Luke, P. Kollias

Pursuing concepts that may include:

• Automatically steering radars to track cells
• Campaign design and processes targeting
• Continual refinement / automation

Multisensor Agile Adaptive Sampling (MAAS)

e.g., Kollias et al. 2020; Lamer et al. 2023

It is critical we maximize chances of observing storms in ways that optimize process-level understanding and lead to accurate predictions.

e.g., Gupta et al., 2024; Giangrande et al. 2023

Capitalize on the influx of open-source “tracking” and instrument simulators.

What I’d Like You to Remember From This Presentation

• ARM deploys a wealth of radars. Each of these is potentially suitable to a variety of applications useful for atmospheric boundary layer, cloud and precipitation studies.

​

• We are still developing best practices for radar to confront many current scientific drivers.

​

• In my experience, to get at the answers I wanted, it often required an understanding for other aspects I did not realize I was interested in.

​

• ARM has a bright future for creative, unique radar-informed boundary layer and cloud process studies.