Francois Cabot

Soil Moisture Retrieval Using Neural Networks: Application to SMOS

A methodology to retrieve soil moisture (SM) from Soil Moisture and Ocean Salinity (SMOS) data is presented. The method uses a neural network (NN) to find the statistical relationship linking the input data to a reference SM data set. The input data are composed of passive microwaves (L-band SMOS brightness temperatures,

SMOS Calibration and Instrument Performance After One Year in Orbit

T_{b}

Remote sensing data respository for in-flight calibration of optical sensors over terrestrial targets

s) complemented with active microwaves (C-band Advanced Scatterometer (ASCAT) backscattering coefficients), and Moderate Resolution Imaging Spectroradiometer (MODIS) Normalized Difference Vegetation Index (NDVI) . The reference SM data used to train the NN are the European Centre For Medium-Range Weather Forecasts model predictions. The best configuration of SMOS data to retrieve SM using an NN is using

Relative and multitemporal calibration of AVHRR, SeaWiFS, and VEGETATION using POLDER characterization of desert sites

T_{b}

Mitigation of RFIS for SMOS: A Distributed Approach

s measured with both H and V polarizations for incidence angles from 25° to 60°. The inversion of SM can be improved by ~10% by adding MODIS NDVI and ASCAT backscattering data and by an additional ~5% by using local information on the maximum and minimum records of SMOS Tbs (or ASCAT backscattering coefficients) and the associated SM values. The NN-inverted SM is able to capture the temporal and spatial variability of the SM reference data set. The temporal variability is better captured when either adding active microwaves or using a local normalization of SMOS Tbs. The NN SM products have been evaluated against in situ measurements, giving results of comparable or better (for some NN configurations) quality to other SM products. The NN used in this paper allows to retrieve SM globally on a daily basis. These results open interesting perspectives such as a near-real-time processor and data assimilation in weather prediction models.

Localization of RFI Sources for the SMOS Mission: A Means for Assessing SMOS Pointing Performances

This paper summarizes the rationale for the European Space Agencys Soil Moisture and Ocean Salinity (SMOS) mission routine calibration plan, including the analysis of the calibration parameter annual variability, and the performances and stability of SMOS images after one year of data. SMOS spends 1.68% of the total observation time in calibration. The instrument performs well within expectations with regard to accuracy and radiometric sensitivity, although spatial ripples are present in SMOS images. Several mechanisms are currently used or under investigation to mitigate this problem. Also, a loss antenna model has recently been introduced to correct for physical temperature-induced effects. This antenna model successfully corrects observed orbital variations, but has difficulties in correcting brightness temperature long-term drifting, as assessed using relatively well-known targets other than the external calibration region-cold space.

A Kurtosis-Based Approach to Detect RFI in SMOS Image Reconstruction Data Processor

The present study is part of an investigation aimed at optimizing the use of desertic sites for absolute or relative calibration of satellite visible sensors. This effort includes characterization of the surface, gathering of climatology or atmospheric data sets, ground- and air- based measurements as well as result of calibration of various sensors over these sites. All these measurements and estimates are stored in a repository and made available to various methods for calibration. Post-launch degradation and relative sensitivity of various sensor have been estimated using north african desertic sites as radiometrically stable targets. The selected area have first been characterize in terms of bidirectional and spectral reflectances by making use of POLDER capabilities, then to cross-calibrate SeaWifs, VEGETATION on-board SPOT4 and AVHRR on-board NOAA-14 by reference to POLDER. Results are compared with absolute and relative calibration issued from other sources. Extensive period of time are spanned to assess the ability of this method to monitor long term trends in sensor evolutions. Results of this cross calibration will be presented. The method developed for this study will be presented as well, in order to make it applicable to other sensor. A sensitivity study has also been realized, considering synthetic data, allowing to evalute the main contributions to the error budget. The need for aerosol optical thickness is then evidenced, and will lead to the set up of a sun photometer on one of the selected sites in 1999.

Soil moisture retrieval from SMOS observations using neural networks

This paper presents the last results of a continuing study aiming at a better characterization of desertic sites for satellite borne optical sensors calibration. The study relies on the gathering of long term archive of satellite data and coincident atmospheric variables, and use of physical properties of the surface as extracted from POLDER measurements, over a set of carefully selected desertic sites in North Africa and Saudi Arabia. All these data sets are included in a repository specifically designed which is also presented. Results are included for AVHRR/NOAA14, SeaWiFS and VEGETATION from 1996 up to present and extension to MODIS and MISR are presented.

An RFI Index to Quantify the Contamination of SMOS Data by Radio-Frequency Interference

The Soil Moisture and Ocean Salinity (SMOS) satellite was launched by the European Space Agency on November 2, 2009. Its payload, i.e., Microwave Imaging Radiometer with Aperture Synthesis, which is a 2-D L-band interferometric radiometer, measures the brightness temperatures (BTs) in the protected 1400-1427-MHz band. Although this band was preserved for passive measurements, numerous radio frequency interferences (RFIs) are clearly visible in SMOS data. One method to get rid of these interferences is to create a synthetic signal as close as possible to the measured interference and subtract it from the instrument visibilities. In this paper, we describe an approach to create such a signal and on how to use it for geolocalization of the emitters. Then, different methods for assessing the quality of the mitigation are introduced. Due to the complexity of estimating the effects of mitigation globally, it is finally proposed to use mitigation results to create flag maps about the estimated RFI impact, to be associated with BT measurements.

Calibration of SMOS geolocation biases

Artificial sources emitting in the protected part of the L-band are contaminating the retrievals of the soil moisture and ocean salinity (SMOS) satellite launched by the European Space Agency (ESA) in November 2009. Detecting and pinpointing such sources is crucial for the improvement of SMOS science products as well as for the identification of the emitters. In this contribution, we present a method to obtain snapshot-wise information about sources of radio-frequency interference (RFI). The localization accuracy of this method is also assessed for observed RFI sources. We also show that RFI localizations constitute a useful data set for assessing the pointing performance of the satellite, and present how it is possible, using the results of this method, to identify and estimate two systematic errors in the geo-location of the satellite field of view. The potential causes and the approaches to mitigate both these errors are discussed.