What Is Radar Polarimetry?

Radar polarimetry is a branch of radar science concerned with measuring and interpreting the polarization of electromagnetic waves scattered by a target or distributed medium. While a conventional intensity-only radar records a single power value per resolution cell, a fully polarimetric radar transmits and receives signals in two orthogonal polarization states, typically horizontal and vertical, and records the full set of complex amplitude ratios between them. This information is encoded in the scattering matrix, a 2x2 complex matrix that characterizes how the target transforms the polarization state of an incident wave into a scattered wave. Analyzing the scattering matrix reveals the geometric and dielectric properties of the reflecting surface or object, providing discrimination capability that intensity alone cannot achieve.

The discipline draws from classical electromagnetic scattering theory, matrix algebra, and statistical signal processing. Its application to spaceborne synthetic aperture radar became practical in the 1980s with the launch of NASA's SIR-C/X-SAR mission, which demonstrated the geophysical value of quad-polarization SAR data for land cover classification and target characterization.

Polarization States and the Scattering Matrix

The polarization of a radar wave describes the orientation and shape of the electric field vector as a function of time at a fixed point in space. Linear polarizations, horizontal (H) and vertical (V), are most commonly used in radar systems; circular polarizations (left and right hand) appear in specialized applications. A monostatic radar operating with reciprocal propagation collects a symmetric scattering matrix with three independent complex entries: HH, VV, and HV (or VH, which equals HV under reciprocity). The magnitude and relative phase of these entries encode information about the target's structure: a smooth horizontal surface returns strong HH and weak HV backscatter, a rough or vegetated surface produces elevated cross-polarization, and a dihedral corner reflector generates a characteristic phase difference between HH and VV. IEEE Xplore publications on target characterization and scattering power decomposition detail how these matrix entries are extracted from fully polarimetric SAR data.

Decomposition Techniques

Polarimetric decomposition methods factor the scattering matrix or its associated covariance matrix into contributions from physically interpretable canonical scatterers. Coherent decompositions, such as the Pauli decomposition, express the scattering matrix as a sum of basis matrices corresponding to odd-bounce reflection, even-bounce reflection, and volume scattering; these are suited to point-like targets. Incoherent decompositions, such as the Freeman-Durden three-component model and the Cloude-Pottier eigenvalue decomposition, operate on the averaged coherency matrix and are used for distributed targets like forest canopies and agricultural fields. The eigenvalue decomposition yields the degree of polarimetric randomness (entropy), the mean scattering angle (alpha), and a relative anisotropy measure that together define a well-established classification space in polarimetric remote sensing. The Springer handbook chapter on basic principles of SAR polarimetry provides a systematic treatment of these decomposition frameworks.

Polarimetric SAR Applications

Polarimetric SAR (PolSAR) systems such as ESA's Sentinel-1 (dual-pol), JAXA's ALOS-2 PALSAR-2 (full-pol), and NASA-ISRO NISAR (full-pol L and S band) provide global, repeat-pass polarimetric data. These systems support retrieval of geophysical parameters including above-ground biomass, forest height, soil moisture, and sea ice type through polarimetric backscatter models calibrated against field observations. Change detection in polarimetric data is sensitive to alterations in surface roughness and vegetation structure, enabling monitoring of deforestation, flood inundation, and urban construction. Research on SAR polarimetry published at the IGARSS symposium series reflects the central role that the IEEE Geoscience and Remote Sensing Society has played in advancing the methodology.

Applications

Radar polarimetry has applications in a wide range of fields, including:

• Agricultural crop classification and growth stage monitoring
• Forest biomass estimation and deforestation detection
• Sea ice type discrimination and ship detection
• Target identification in military and security contexts
• Urban land use mapping and building damage assessment
• Soil moisture and surface roughness retrieval