Faraday-Rotation Prediction Values on URSA

by Franz Meyer, Alaska Satellite Facility

WHAT IS FARADAY ROTATION?

SAR signals that propagate through the ionosphere undergo Faraday rotation (FR), i.e., a rotation of its polarization vector. After traversing the ionosphere, vertically polarized signals exhibit both vertical and horizontal components, and received SAR data will contain cross-pol terms.

FR is frequency dependent and increases with decreasing signal frequency. Therefore, data from X- and C-band SARs are practically unaffected, while effects on L-band SARs (e.g., ALOS Phased Array type L-band Synthetic Aperture Radar (PALSAR) can be significant. Besides the frequency dependence, the magnitude of FR depends primarily on the following two parameters:

• 1. The free-electron density of the ionosphere (cf. Figure 1) during acquisition; mainly influenced by the solar activity, the geographic location, and the time of day;
• 2. the magnitude and orientation of the Earth’s magnetic field.

FR generally reduces the image quality of SAR images acquired in L-band and will impact ALOS PALSAR imagery. While FR angles of less than 5º were established as acceptable for a number of common SAR applications, higher FR levels may cause significant errors in SAR image interpretation and data analysis, especially decompositions relying on channel ratios.

FR PREDICTION

FR angles can be predicted if information about satellite orbit, ionospheric conditions, and the magnetic field are available. On the User Remote Sensing. Access (URSA) order interface, we provide customers with a listing of predicted FRs for PALSAR acquisitions in the ASF ALOS archive. This is intended to aid data selection for researchers wishing to avoid FR effects, or for those scientists deliberately targeting these effects.

As is true for every predicted parameter, a given FR prediction is only an approximation of its true value. An analysis of the prediction quality yields an accuracy of about 1º. Inaccuracies stem from uncertainties and, low spatial and temporal resolution of the input data required by the model.

When searching for data on the URSA order interface, customers can specify the acceptable range of FR and will be provided with a selection of datasets that optimally meet their requirements.

Click here to download a copy of the newsletter featuring this article