Meetings: Documents

Comparison of L-band Radio Frequency Interferences from Aquarius and SMOS Observations
[11-Nov-2014] Soldo, Y., de Matthaeis, P., Le Vine, D., and Richaume, P.
Presented at the 2014 Aquarius/SAC-D Science Team Meeting
The ESA (European Space Agency) SMOS (Soil Moisture and Ocean Salinity) satellite and the NASA (National Aeronautics and Space Administration) Aquarius instrument have been measuring Earth's natural emissions in the protected part of the L-band (1400-1427 MHz) since January 2010 and August 2011, respectively. SMOS is an interferometric radiometer designed to retrieve surface salinity over oceans and moisture over landmasses, whereas Aquarius is composed by three real-aperture radiometers and its main purpose is the retrieval of sea surface salinity.
Even though the spectral band used by the two instruments has been preserved for passive measurements by ITU (International Telecommunication Union) regulations, both instruments see the quality of their observations degraded by significant Radio Frequency Interference (RFI), especially over land.
In this work, the RFI contamination in both instruments have been compared through the "RFI probability" associated to points on Earth's surface, i.e. the ratio between the number of times RFI were detected on a certain point and the number of times the same point was observed. The comparison of RFI probabilities from both instruments has been done: 1) on a single, isolated and strong source, and 2) globally, over one month.
In both instruments, the contamination by RFI sources originates in the same way: the energy coming from artificial sources emitting in the protected band reaches the instrument, and the instruments registers brightness temperatures higher than the expected natural thermal noise. But, of course, there are also several differences in the way Aquarius and SMOS are affected by the RFI sources. These differences include: the times of observation of the same sources, the spectral bandwidths used, the antenna directivities, the size of the radiometric pixels, the capability of making multi-angular observations, the acquisition sequences and the approaches used to detect RFI.
Because of these differences, the RFI experienced by both instruments are compared not only through the nominal RFI probabilities, that can be obtained from both instruments independently, but also with a more elaborate approach that aims at making Aquarius and SMOS data as similar as possible, i.e. addressing the differences listed above. In this approach the Aquarius data is projected onto the Discrete Global Grid (DGG) used in SMOS Level-1C and the SMOS field of view is matched to the Aquarius beams.
This comparison shows generally a good agreement between the two sets of RFI probabilities, but also some cases in which the instruments give very different results (Japan and France). It also shows how SMOS suffers because of its wider field of view, while Aquarius suffers because of the wider bandwidth it uses.
Another approach has been to consider only the RFI sources that are always detected by one of the two instruments and to assess the corresponding detections by the other instrument. The results of this approach show that high RFI probabilities in SMOS correspond to high RFI probabilities in Aquarius, though not conversely, because of the higher spatial heterogeneity of SMOS RFI probabilities.