Active Remote Sensing and Explorations (part 1: 2.1–2.2)
Open in the Padhaiko app2.1 Principles of Microwave Remote Sensing
Unlike passive optical sensors that rely on reflected sunlight or thermal self-emission, active microwave remote sensing systems (radars) transmit their own coherent electromagnetic pulses in the () band and measure the backscattered echo returned by the target. Because long microwave wavelengths are minimally attenuated by clouds, haze, and light rain and operate independently of solar illumination, radar provides all-weather, day-and-night Earth observation.
| Band | Frequency Range | Typical Wavelength () | Primary Applications |
|---|---|---|---|
| P-band | Deep forest biomass, sub-canopy & dry-soil penetration (BIOMASS) | ||
| L-band | Forestry, soil moisture, crust deformation (ALOS-2/4, NISAR) | ||
| S-band | Weather/ATC radar, NISAR S-band SAR, agriculture | ||
| C-band | General land/sea monitoring, floods, oil spills (Sentinel-1, RADARSAT) | ||
| X-band | High-resolution urban mapping, surveillance, marine radar (TerraSAR-X) | ||
| Ku / Ka-band | Radar altimetry, snow/ice surface scattering, precipitation radar |
Microwave Interaction with Surfaces
When a microwave wavefront strikes a terrain or target surface, the strength of the backscattered signal (quantified by the normalized radar cross section or backscatter coefficient , often expressed in as ) is governed primarily by two groups of properties:
- Target properties: Complex dielectric constant (moisture content, salinity, material type) and surface roughness / geometry (micro-roughness, vegetation structure, corner reflectors).
- Sensor / viewing properties: Wavelength (), polarization, and local incidence angle ().
Dielectric Properties
The electrical response of a target medium to an oscillating microwave field is characterized by its complex relative permittivity (dielectric constant):
- The real part governs wave velocity () and surface Fresnel reflectivity (how much energy reflects rather than transmits).
- The imaginary part (loss factor, related to conductivity ) governs dielectric absorption (attenuation) inside the medium.
Polarization
Polarization specifies the orientation of the oscillating electric field vector in the plane perpendicular to propagation:
- Horizontal polarization (H): oscillates parallel to the Earth's horizontal surface.
- Vertical polarization (V): oscillates in the vertical plane containing the radar look vector.
Radar systems transmit and receive in four linear combinations:
- Co-polarized (Like-polarized): HH (transmit H, receive H) and VV (transmit V, receive V). Sensitive to single-bounce surface scattering and double-bounce trunk–ground corners (especially HH).
- Cross-polarized: HV and VH. Generated by volume scattering inside vegetation canopies or multiple reflections that rotate the polarization plane; strongly sensitive to forest biomass and crop structure.
Incidence Angle
The local incidence angle is the angle between the incoming radar beam and the normal to the illuminated surface element:
- Near-nadir / steep incidence (): Specular reflection dominates on smooth surfaces, while radar foreshortening and layover distort steep terrain slopes.
- Moderate incidence (): Optimal compromise for terrain backscatter sensitivity and geometric fidelity; smooth water reflects energy away (appears dark), whereas rough terrain scatters back strongly.
- Grazing incidence (): Backscatter drops rapidly on flat surfaces, and tall terrain ridges cast radar shadows where no signal returns.
Penetration Depth
The penetration depth (skin depth ) is the depth below a surface at which the incident microwave power decays to () of its surface value (or electric field amplitude decays to , with power skin depth ):
Longer wavelengths (L-band, P-band) and dry, low-loss materials (hyper-arid desert sand, dry snow, cold ice) permit meters of subsurface or canopy penetration, whereas wet soils and liquid water limit C- and X-band penetration to a few millimeters or centimeters.
Surface Roughness
Whether a surface acts as a smooth (specular) reflector or a rough (diffuse) backscatterer depends on the root-mean-square (RMS) surface height variation relative to and .
Rayleigh and Fraunhofer Surface Roughness Criteria
Let be the RMS height deviation of surface irregularities, the radar wavelength, and the incidence angle:
- Rayleigh Criterion: A surface is considered smooth if the phase difference between waves reflected from high and low surface points satisfies , yielding:
- Fraunhofer Criterion (stricter, ): Distinguishes truly smooth from intermediate and rough surfaces:
- Smooth: (specular reflection away from the radar dark tone in SAR).
- Rough: (diffuse scattering in all directions bright backscatter).
- Intermediate: Between and .
2.2 Basic Radar Range Equations, Parameters, Systems and Indicators
The Radar Range Equation
Consider a monostatic radar (collocated transmitter and receiver) radiating peak power through an antenna of gain toward a target at slant range with radar cross section (RCS) ().
Monostatic Radar Range Equation
- Power density at the target: An isotropic radiator spreads over a sphere of area ; antenna gain focuses power to
- Reradiated power density back at the radar: A target of radar cross section intercepts power and scatters it isotropically back over distance :
- Received power (): Captured by an antenna of effective aperture , accounting for two-way system/propagation losses :
- Maximum Detection Range (): Setting (the minimum detectable signal power) yields:
Notice the fourth-power dependence () of received power on range: doubling the target range reduces the received echo power by a factor of (), and doubling the range requires increasing transmit power by a factor of .
Range and Resolution
A pulsed radar determines slant range from the two-way round-trip travel time of the electromagnetic pulse:
The factor of accounts for the outbound and return path.
Slant-Range, Ground-Range, and Angular Resolution
- Slant-Range Resolution (): Two targets along the line of sight are separable only if their echoes do not overlap in time, requiring a path difference of at least :
where is the effective (compressed) pulse duration and is the radar signal bandwidth.
- Ground-Range Resolution (): Projecting slant range onto horizontal terrain at incidence angle :
- Real-Aperture Azimuth Resolution (): Determined by the antenna horizontal beamwidth :
where is the physical antenna length along track. (For spaceborne , real-aperture azimuth resolution would be kilometers wide, motivating Synthetic Aperture Radar in §2.3.6.)
Pulse Duration, Bandwidth, and Pulse Repetition Frequency (PRF)
- Pulse Duration (): Shorter pulses yield finer range resolution (), but carry less total energy for a fixed peak power . Modern radars overcome this trade-off using pulse compression (linear FM chirp): transmitting a long pulse of duration swept across wide bandwidth , then matched-filtering the echo to achieve an effective compressed width (time–bandwidth product ).
- Pulse Repetition Frequency ( or ) and Duty Cycle: The radar transmits pulses at rate , where is the Pulse Repetition Interval (PRI). The duty cycle is , and average power is .
- Maximum Unambiguous Range (): An echo from a pulse must return before the next pulse is transmitted () to avoid range ambiguity (second-time-around echoes):
Increasing PRF improves Doppler/azimuth sampling but reduces .
Antenna Gain, Wavelength, and Bandwidth
- Antenna Gain (): Ratio of radiation intensity in the peak beam direction to that of an isotropic radiator. For an antenna of physical aperture and aperture efficiency :
Approximated in terms of azimuth () and elevation () 3-dB beamwidths (in radians) as .
- Wavelength (): Shorter wavelengths yield higher gain for a given dish size and finer real-aperture angular resolution, but suffer greater atmospheric/rain attenuation and reduced canopy/soil penetration.
- Bandwidth (): Wider receiver/chirp bandwidth directly improves slant-range resolution (), while increasing thermal noise power .
Signal-to-Noise Ratio (SNR) and System Indicators
Receiver thermal noise power over bandwidth is , where , is standard reference temperature, and is the receiver noise figure (linear ratio). Substituting from the radar equation gives the single-pulse SNR:
Key system performance indicators derived from SNR include:
- Minimum Detectable Signal (): , where satisfies required detection probability () and false-alarm probability ().
- Noise Equivalent Sigma Zero ( or ): The surface backscatter coefficient that produces (). Lower (more negative in dB, e.g., ) NESZ indicates higher radiometric sensitivity over dark surfaces like calm water.
Worked Out Examples
2.1
A radar observes an agricultural field at incidence angle using two sensors: a C-band SAR () and an L-band SAR ().
- Using the Fraunhofer criterion, find the smooth and rough RMS height thresholds ( in cm) for the C-band sensor.
- Using the Rayleigh criterion, find the rough threshold for the L-band sensor, and classify a tilled field with in both bands.
Applying the Fraunhofer thresholds for C-band (, ): First compute :
- Smooth limit:
- Rough limit:
Since , the tilled field appears rough (bright diffuse scatterer) at C-band.
Applying the Rayleigh threshold for L-band (, ):
Notice the Fraunhofer smooth threshold at L-band is . Because , the tilled field satisfies the Rayleigh smoothness condition () and lies in the intermediate regime under Fraunhofer (appearing much smoother/darker at L-band than at C-band).
- C-band Fraunhofer thresholds: smooth if , rough if .
- L-band Rayleigh threshold: . The field is rough at C-band but smooth (Rayleigh) / intermediate (Fraunhofer) at L-band.
2.2
A pulsed X-band radar operates with pulse repetition frequency , chirp bandwidth , and off-nadir incidence angle . Calculate:
- The maximum unambiguous range in km.
- The slant-range resolution .
- The ground-range resolution at .
Computing maximum unambiguous range:
Computing slant-range and ground-range resolution:
Projecting onto the horizontal ground at ():
- .
- .
- .
2.3
A ground radar at transmits peak power through an antenna of gain () toward a target of radar cross section at range . System losses are , receiver bandwidth is , and receiver noise figure is at .
- Compute the received echo power in Watts.
- Compute the single-pulse signal-to-noise ratio () in linear units and in .
Converting dB quantities to linear ratios:
Evaluating the radar range equation for :
Evaluating numerator and denominator:
- Numerator:
- Denominator:
Computing thermal noise power and single-pulse SNR:
Converting to decibels:
- Received echo power: .
- Single-pulse ().
Exercises
Exercise — Principles of Microwave Remote Sensing & Radar Equation
Explain how the complex dielectric constant (), polarization (HH, VV, HV, VH), and local incidence angle () govern microwave backscatter from soil and vegetation surfaces.
Role of complex dielectric constant: controls surface reflectivity and wave speed (), while governs subsurface attenuation. Because liquid water has compared to for dry soil and ice, increases in soil or canopy moisture sharply raise and reduce penetration depth.
Role of polarization: Co-polarized channels (HH, VV) respond strongly to surface scattering and trunk–ground double-bounce (especially HH), whereas cross-polarized channels (HV, VH) arise from multiple volume scattering inside vegetation canopies, making HV/VH sensitive to biomass.
Role of incidence angle: Steep incidence angles () emphasize specular and topographic foreshortening/layover effects; moderate angles () balance surface roughness sensitivity and geometry; high angles () reduce flat-surface return and induce radar shadows behind steep terrain.
High water dielectric constant () amplifies backscatter and limits penetration; HH/VV capture surface and double-bounce scattering while HV/VH capture canopy volume scattering; trades foreshortening/layover at steep angles against shadowing at grazing angles.
Define penetration depth () in microwave remote sensing and explain why P- and L-band radars penetrate dry sand or forest canopies much deeper than X-band radars. Estimate the power penetration depth in dry desert sand (, ) at L-band ().
Definition and wavelength/loss dependence: Penetration depth is the depth where microwave power drops to of its surface value: for low-loss media (). Longer wavelengths () and low-loss dry materials (small ) increase linearly with .
Numerical calculation for dry sand at L-band:
is the power attenuation depth; larger and smaller yield deeper penetration. In dry sand () at , .
State the Rayleigh and Fraunhofer surface roughness criteria. An X-band radar () illuminates a road and a gravel surface at . Compute the Rayleigh RMS height threshold in cm, and explain why calm water appears dark on SAR imagery while rough gravel appears bright.
Rayleigh and Fraunhofer roughness criteria:
- Rayleigh smooth: .
- Fraunhofer smooth: ; rough: .
Numerical threshold at :
Physical interpretation: Calm water () acts as a specular reflector, bouncing incident radar energy forward away from the monostatic receiver so almost no power returns (dark pixel). Rough gravel () scatters energy diffusely in all directions, returning significant backscatter toward the radar (bright pixel).
at . Calm water reflects specularly away from the sensor (dark), whereas rough gravel scatters diffusely back to the sensor (bright).
Derive the monostatic radar range equation . A radar transmits unmodulated pulses of duration at and incidence angle . Find (a) slant-range resolution , (b) ground-range resolution , and (c) maximum unambiguous range .
Derivation of the radar range equation: Transmitted power density at range is . A target of RCS reradiates over , giving return flux . Multiplying by effective antenna aperture and dividing by loss yields .
Computing resolutions and unambiguous range:
- (a)
- (b)
- (c)
(a) ; (b) ; (c) .
Explain the trade-off between pulse duration , bandwidth , PRF, antenna gain , and SNR in radar design, and describe how pulse compression resolves the range-resolution vs. energy dilemma. An airborne C-band radar () uses an antenna of physical area with efficiency , pulse duration , and . Calculate (a) the linear antenna gain and (b) the radar duty cycle.
Pulse compression and parameter trade-offs: Short unmodulated pulses improve range resolution () but carry low pulse energy , degrading SNR; wide receiver bandwidth improves resolution () while admitting more thermal noise . Pulse compression transmits a long frequency-modulated chirp (high energy) across wide bandwidth and compresses it via matched filtering to . Higher PRF improves sampling but lowers .
Computing antenna gain and duty cycle:
- (a) Physical area :
- (b) Duty cycle:
Pulse compression decouples pulse energy (long ) from range resolution (wide ). For the given radar: (a) (); (b) duty cycle ().