Compact Polarimetry
Understanding Compact Polarimetry
Full polarimetric (quad-pol) synthetic aperture radar measures the complete 2x2 complex scattering matrix by alternately transmitting horizontal (H) and vertical (V) polarizations and receiving both on each pulse. That measurement is information-rich, but it carries a heavy cost: the radar must transmit two polarizations, so its effective pulse repetition frequency (PRF) requirement doubles. Doubling the PRF halves the unambiguous swath width and doubles the raw data volume that must be stored and downlinked. For wide-area operational mapping (sea-ice charting, crop classification, disaster response, biomass estimation) a 25 to 30 km quad-pol swath is often too narrow to be useful. Compact polarimetry was developed to recover most of the polarimetric value while keeping the wide swath and modest data rate of a single-transmit system.
The central idea is to transmit one carefully chosen polarization and to coherently record two orthogonal receive channels, keeping their amplitude and, critically, their relative phase. Two complex receive numbers per resolution cell are enough to populate the four real-valued Stokes parameters, which fully describe the partially polarized backscattered wave. From the Stokes vector, analysts derive familiar quantities such as the degree of polarization, the ellipticity and orientation angles, and the relative phase between channels. These feed decomposition methods (the m-delta, m-chi, and m-alpha decompositions) that separate surface, double-bounce, and volume scattering much as full-pol decompositions do.
Transmit and Receive Architectures
Three compact configurations are used in practice. The pi/4 mode transmits a 45-degree linear polarization (an equal, in-phase combination of H and V) and receives H and V. The dual circular polarimetric (DCP) mode transmits circular polarization and receives left and right circular. The circular transmit, linear receive (CTLR) mode, also called hybrid-polarity, transmits circular and receives H and V. CTLR has become the de facto standard for spaceborne systems because the circular transmit state is the average of two orthogonal linear states, which makes the mode insensitive to the orientation of the scene relative to the antenna and, importantly, far more tolerant of Faraday rotation in the ionosphere at L-band and lower frequencies.
Reconstruction and Its Assumptions
A compact-pol system measures two of the three independent terms of the reciprocal scattering matrix; it does not directly capture the full matrix. To estimate pseudo-quad-pol covariance terms, reconstruction algorithms invoke assumptions about the scene, most commonly reflection symmetry (the cross-correlation between co-polarized and cross-polarized returns averages to zero over a distributed target). These assumptions hold well for many natural surfaces such as forests, agricultural fields, and ocean, but they degrade over strongly oriented or man-made structures. Engineers therefore treat reconstructed quad-pol products as approximations and often prefer to work directly in the Stokes or covariance domain where no symmetry assumption is required.
Mission Heritage and Applications
Compact polarimetry has flown or is planned on several operational missions, including India's RISAT-1, the RADARSAT Constellation Mission, ALOS-2 experimental modes, and the L-band/S-band NASA-ISRO NISAR mission. Typical applications include sea-ice type and concentration mapping, oil-spill detection, soil-moisture and crop-stage classification, flood and wetland delineation, and forest disturbance monitoring. The wide swath that compact polarimetry enables is the deciding advantage: a single pass can cover several hundred kilometers while still delivering polarimetric discrimination that intensity-only dual-pol cannot provide.
Key Equations
g0 = ⟨|EH|²⟩ + ⟨|EV|²⟩
g1 = ⟨|EH|²⟩ − ⟨|EV|²⟩
g2 = 2·Re⟨EH EV*⟩
g3 = −2·Im⟨EH EV*⟩
Degree of polarization:
m = √(g1² + g2² + g3²) / g0
Where g0 = total received power, g1 = H/V power imbalance, g2 and g3 = in-phase and quadrature correlation of the two channels, Re/Im = real and imaginary parts, ⟨·⟩ = spatial (multi-look) ensemble average, E* = complex conjugate, and 0 ≤ m ≤ 1. The relative phase delta = arctan(g3/g2) drives the m-delta surface/double-bounce decomposition.
SAR Polarimetric Mode Comparison
| Mode | Transmit | Receive | Relative PRF | Swath (relative) | Scattering Info |
|---|---|---|---|---|---|
| Single-pol | 1 (e.g. H) | 1 (e.g. H) | 1x | Widest | Intensity only |
| Dual-pol (intensity) | 1 (e.g. H) | 2 (HH, HV) | 1x | Wide | Two intensities, no phase |
| Compact-pol (CTLR) | 1 (circular) | 2 (H, V, coherent) | 1x | Wide | Stokes vector, pseudo quad-pol |
| Compact-pol (pi/4) | 1 (45° linear) | 2 (H, V, coherent) | 1x | Wide | Stokes vector, less Faraday-robust |
| Quad-pol (full) | 2 (H and V) | 2 (H, V) | 2x | ~Half | Complete 2x2 scattering matrix |
Frequently Asked Questions
What is compact polarimetry?
Compact polarimetry is a dual-polarized synthetic aperture radar mode that transmits one fixed polarization state, usually circular or 45-degree linear, and coherently receives two orthogonal linear polarizations while preserving their relative phase. From these two complex channels the system reconstructs the four-element Stokes vector and pseudo-quad-pol scattering descriptors. This approximates the information content of a full quad-pol radar while using about half the pulse repetition frequency, which doubles the achievable swath width and roughly halves the downlink data rate. The penalty is that the full 2x2 scattering matrix is not directly measured, so reciprocity and scene assumptions are needed during reconstruction.
What are the main compact polarimetry modes?
Three architectures dominate. The pi/4 mode transmits 45-degree linear polarization and receives H and V. The dual circular polarimetric (DCP) mode transmits circular and receives left and right circular. The circular transmit, linear receive (CTLR) or hybrid-polarity mode transmits circular and receives H and V; CTLR is the most widely deployed because it relaxes transmit-channel calibration and is robust to Faraday rotation at L-band. Spaceborne examples include the Indian RISAT-1, NASA-ISRO NISAR, Canada's RADARSAT Constellation Mission, and the Japanese ALOS-2 experimental modes.
How does compact polarimetry compare to quad-pol and dual-pol SAR?
Quad-pol (fully polarimetric) SAR transmits H and V on alternate pulses and measures the complete 2x2 complex scattering matrix, but it requires roughly double the pulse repetition frequency, which halves the unambiguous swath and doubles the data volume. Conventional dual-pol records two intensity channels such as HH and HV but loses the relative phase, so it cannot reconstruct polarimetric decompositions. Compact polarimetry sits between them: it keeps the relative phase of two receive channels, recovers the Stokes parameters and useful decomposition information, yet retains the wide swath and low data rate of a single-transmit system. It is therefore favored for operational wide-area mapping where full quad-pol coverage would be too narrow.