Soil moisture content (SMC) and above-ground biomass (AGB) are key parameters for the understanding of both the hydrological and carbon cycles. From an economical perspective, both SMC and AGB play a significant role in the agricultural sector, one of the most relevant markets worldwide. This paper assesses the sensitivity of Global Navigation Satellite System (GNSS) reflected signals to soil moisture and vegetation biomass from an experimental point of view. For that, three scientific flights were performed in order to acquire GNSS reflectometry (GNSS-R) polarimetric observations over a wide range of terrain conditions. The GNSS-R data were used to obtain the right-left and right-right reflectivity components, which were then georeferenced according to the transmitting GNSS satellite and receiver positions. It was determined that for low-altitude GNSS-R airborne platforms, the reflectivity polarization ratio provides a highly reliable observable for SMC due to its high stability with respect to surface roughness. A correlation coefficient r2 of 0.93 and a sensitivity of 0.2 dB/SMC (%) were obtained for moderately vegetated fields with a surface roughness standard deviation below 3 cm. Similarly, the copolarized reflection coefficient shows a stable sensitivity to forest AGB with r2 equal to 0.9 with a stable sensitivity of 1.5 dB/(100 t/ha) up to AGB values not detectable by other remote sensing systems. © 2014 IEEE

No abstract

No abstract

Abstract:The work focuses on developing a classification tree approach for in-season crop mapping during early summer, by integrating optical (Landsat 8 OLI) and X-band SAR (COSMO-SkyMed) data acquired over a test site in Northern Italy. The approach is based on a classification tree scheme fed with a set of synoptic seasonal features (minimum, maximum and average, computed over the multi-temporal datasets) derived from vegetation and soil condition proxies for optical (three spectral indices) and X-band SAR (backscatter) data. Best performing input features were selected based on crop type separability and preliminary classification tests. The final outputs are crop maps identifying seven crop types, delivered during the early growing season (mid-July). Validation was carried out for two seasons (2013 and 2014), achieving overall accuracy greater than 86%. Results highlighted the contribution of the X-band backscatter (σ°) in improving mapping accuracy and promoting the transferability of the algorithm over a different year, when compared to using only optical features.

Remote sensing of vegetation by using active microwave sensors is important for the management of land and water resources. Microwave radiation at X-band penetrates only the upper part of the canopy; thus, radar backscattering comes mainly from the top vegetation layer, making the scattering from soil almost negligible. Fourteen in-situ measurement campaigns were carried out during which sixteen SAR images of COSMO-SkyMed and TerraSAR-X were acquired on the test site of Merguellil Basin, in the center of Tunisia, from March to May 2012. A clear sensitivity of the backscattering coefficient, measured by both sensors, to the leaf area index (LAI) of green plants of wheat and barley (at both HH-and VVpolarizations) was observed, and it did not seem to be greatly affected by the variations in soil moisture, even in HH polarization. As expected, the sensitivity to LAI of dry plants was, instead, almost negligible. The decreasing trend of backscattering as vegetation grows, which has already been observed in past investigations, also at lower frequencies, was confirmed. Due to the similar geometry and dimensions of the two crop types, none of the sensors succeeded in separating the two crop types.

In this study, the capability of Global Navigation Satellite System Reflectometry in evaluating forest biomass from space has been investigated by using data coming from the TechDemoSat-1 (TDS-1) mission of Surrey Satellite Technology Ltd. and from the Cyclone Satellite System (CyGNSS) mission of NASA. The analysis has been first conducted using TDS-1 data on a local scale, by selecting five test areas located in different parts of the Earth's surface. The areas were chosen as examples of various forest coverages, including equatorial and boreal forests. Then, the analysis has been extended by using CyGNSS to a global scale, including any type of forest coverage. The peak of the Delay Doppler Map calibrated to retrieve an "equivalent" reflectivity has been exploited for this investigation and its sensitivity to forest parameters has been evaluated by a direct comparison with vegetation optical depth (VOD) derived from the Soil Moisture Active Passive L-band radiometer, with a pantropical aboveground biomass (AGB) map and then with a tree height (H) global map derived from the Geoscience Laser Altimeter System installed on-board the ICEsat satellite. The sensitivity analysis confirmed the decreasing trend of the observed equivalent reflectivity for increasing biomass, with correlation coefficients 0.31 ≤ R ≤ 0.54 depending on the target parameter (VOD, AGB, or H) and on the considered dataset (local or global). These correlations were not sufficient to retrieve the target parameters by simple inversion of the direct relationships. The retrieval has been therefore based on Artificial Neural Networks making it possible to add other inputs (e.g., the incidence angle, the signal to noise ratio, and the lat/lon information in case of global maps) to the algorithm. Although not directly correlated to the biomass, these inputs helped in improving the retrieval accuracy. The algorithm was tested on both the selected areas and globally,

No abstract

In this paper, multifrequency synthetic aperture radar (SAR) images from ALOS/PALSAR, ENVISAT/ASAR and Cosmo-SkyMed sensors were studied for forest classification in a test area in Central Italy (San Rossore), where detailed in-situ measurements were available. A preliminary discrimination of the main land cover classes and forest types was carried out by exploiting the synergy among L-, C- and X-bands and different polarizations. SAR data were preliminarily inspected to assess the capabilities of discriminating forest from non-forest and separating broadleaf from coniferous forests. The temporal average backscattering coefficient ( σ ¯ °) was computed for each sensor-polarization pair and labeled on a pixel basis according to the reference map. Several classification methods based on the machine learning framework were applied and validated considering different features, in order to highlight the contribution of bands and polarizations, as well as to assess the classifiers’ performance. The experimental results indicate that the different surface types are best identified by using all bands, followed by joint L- and X-bands. In the former case, the best overall average accuracy (83.1%) is achieved by random forest classification. Finally, the classification maps on class edges are discussed to highlight the misclassification errors.