Compensating PALSAR Imagery for Ionosphere Effects

Jong-Sen Lee(1), Thomas Ainsworth(2) and Kun-Shan Chen(3)

(1) National Central University, Chung-Da Road, Chung-Li, Taiwan, Other
(2) Naval Research Laboratory, 4555, Overlook Ave. SW, Washington DC, 20375, United States
(3) CSRSR, Chung Da Road, Chung-Li, Taiwan, Other


Compensating PALSAR Imagery for Ionosphere Effects

PI: Jong-Sen Lee, PI Number: 317 – JAXA-RA2

T. L. Ainsworth1, J.S. Lee2 and K.S. Chen2

1Remote Sensing Division, Naval Research Lab., Washington, DC 2CSRSR, Central National University, Taiwan

Correction of Faraday rotation effects in polarimetric SAR data can in principle present difficult problems. PALSAR microwave signals traverse the Earth’s ionosphere twice. Thus the radar signals are affected by the local properties of the ionosphere. Two critical properties of the local ionosphere are the free electron density and the direction of the Earth’s magnetic field with respect to the radar propagation direction. The Faraday rotations are proportional to both the integrated free electron density along the path of the radar signal, the total electron content (TEC) of the ionosphere, and the component of the magnetic field along the propagation path. The strength and direction of the Earth’s magnetic field remains fairly constant over the length scales of SAR images, say 50km, except possibly near the magnetic poles where the field lines are strongly curved. However, TEC values can vary on length scales much smaller than the SAR image. Therefore, localized inhomogeneities in the TEC cause Faraday distortions which vary across the image.

Correcting for a uniform Faraday rotation across a SAR image is fundamentally similar to polarimetric calibration. General polarimetric SAR calibration methods depend both on the linearization of the non-linear calibration equations and, in practice, on the small magnitude of the crosstalk and channel imbalance corrections. We first present the full non-linear solution to polarimetric calibration in the presence of Faraday rotations and then investigate approximate calibration methods that assume either zero cross-talk between polarimetric channels or a reciprocal radar system. We will test the Faraday correction and polarimetric calibration procedures using both PALSAR imagery and simulated datasets that show effects due to localized TEC inhomgeneities.

The TEC primarily affects the polarization state of the radar signals traversing the ionosphere. However small scale inhomogeneities also generate scintillation effects which then distort SAR image formation. We will present a preliminary assessment of PALSAR image distortions that may arise from these small-scale equatorial ionospheric disturbances. .


Symposium presentation


  Higher level                 Last modified: 07.05.06