Rachid Rahmoune

SMOS Level 2 Retrieval Algorithm Over Forests: Description and Generation of Global Maps

This paper shows global maps of optical depth and soil moisture over land, obtained using the last prototype of SMOS Level 2 retrieval algorithm, which will be implemented in V600 version of Level 2 product made available by European Space Agency (ESA). The focus is on forested areas, where the approach adopted to develop the algorithm can be subdivided into different steps. First a theoretical model, which was previously developed and tested using ground based and airborne measurements, generated parametric outputs. By fitting this output data set, the albedo and the optical depth of a simple first order radiative transfer model were estimated. Then, this simplified forest model was included in the general ESA Level 2 retrieval algorithm over land, described in the Algorithm Theroretical Baseline Document (ATBD). The paper describes the details of this procedure and shows some retrieval results. First, the prototype algorithm was run with three free parameters: Soil moisture, optical depth, and albedo. The retrieved albedo resulted to be close to the initial estimate (0.08) for Boreal forests, while it was lower for Tropical forests. Running again the algorithm with the albedo fixed, a global map of optical depth was generated. The spatial features of the map follow the global information about forest biomass and forest height available in the literature. Finally it was found that, on average, the influence of seasonal effects on optical depth is moderate.

Modeling the Multifrequency Emission of Broadleaf Forests and Their Components

This paper shows a model study about the emissivity of forests. Model outputs are compared with multifrequency airborne measurements carried out over five broadleaf forests in Italy. Two flights took place, in summer 1999 and winter 2002. Available ground truth included important variables, such as biomass, tree density, and average trunk diameter. This data set, in conjunction with allometric equations and information taken from the literature, is used to give inputs to the model. A general agreement between simulated and measured data is observed at L-, C-, and X-bands. The same model is used to investigate the sensitivity of forest emissivity to soil moisture, woody volume, and average diameter. As expected, a moderate effect of soil moisture is observed only at L-band and for forests with a lower woody volume. At L-band, the model predicts a general increase of emissivity with woody volume but indicates that also the trunk diameter exerts an important influence, since it is a variable which controls several geometrical properties. These results allow us to single out the influence of soil moisture, woody volume, and geometrical properties at L-band. The increase of emissivity with frequency, observed in experimental data, is interpreted by means of electromagnetic considerations about branch scattering.

SMOS Retrieval Results Over Forests: Comparisons With Independent Measurements

This paper shows results obtained by using SMOS Level 2 retrieval algorithm, run at prototype stage, over forests. For each SMOS pixel, the algorithm estimates the soil moisture (SM) and the vegetation optical depth (τ). Average τ values retrieved in 4 days of July 2011 over forest pixels are shown and compared against forest height estimated by GLAS Lidar on board ICEsat satellite. Results of the comparison show a significantly increasing trend of τ with respect to forest height. For each 1-m interval of forest height estimated by Lidar, the standard deviation of optical depth is slightly higher than 0.1. The analysis is made again considering forest τ retrieved in 4 days of February, May, and November 2011, and it is observed that seasonal effects over optical depth are low. As an insight, it is shown that the increasing trend is still observed after subdividing world forests into Coniferous, Deciduous Broadleaf, and Evergreen Broadleaf. Comparisons with independent information about biomass are also shown at regional level for the U.S. The increasing trend is still observed, but with a reduced range of values. For SM, 14 nodes of the SCAN/SNOTEL network in the U.S. are considered. Over 2 years of data, retrieved values of SM are compared against ground measurements. Obtained values of correlation coefficient, rms error, and bias are reported.

The Effect of Rain and Flooding Events on AMSR-E Signatures of La Plata Basin, Argentina

The objective of this paper is to describe and explain the effects on selected AMSR-E channels of two strong events, i.e., a rainstorm and a flooding, occurred in the Argentine section of La Plata basin. More specifically, the rainstorm took place within the Chaco region, which is covered by a continuous, moderately dense forest. The flooding affected the terminal part of Parana¿ River. The study is based on monitoring the temporal trends of the polarization indexes at various AMSR-E bands. In the forest, the rainstorm produces an effect on C band channels which is moderate, but well evident. The presence of this effect agrees with model simulations presented in previous papers. In the Parana¿ River, measurements of water level are available. Variations of polarization index at various frequencies are observed in correspondence with variations of water level in four different stations. However, the amount of the effect and the correlation between variables are dependent on the properties of the areas surrounding the stations. The Delta of Parana¿ river, where a land cover map is available, was selected for estimation of fraction of flooded area by using an algorithm available in the literature.

Flood monitoring using microwave passive remote sensing AMSR-E in part of the Brahmaputra basin, India

In this work, Advanced Microwave Scanning Radiometer Earth Observing System (AMSR-E) signatures were analysed over some critical sites in Lakhimpur District in Brahmaputra basin, India, characterized by a high frequency of flooding events. The site is mostly covered by paddy fields. Results obtained were compared with water level measurements in three stations close to the main channel of the river. Information about surface temperature, which allowed us to estimate the emissivity, was also available. Investigations were carried out at the C, X, and Ka bands of the AMSR-E channel. A multi-frequency analysis indicated that the X band would represent a good compromise between resolution and sensitivity requirements, while at the C band the resolution was too coarse and at the Ka band the signatures were affected by raindrops. Samples collected during rain were eliminated using techniques based on the 89.0 GHz channel. However, even after this correction, the Ka band showed poor sensitivity due to higher attenuation by vegetation. The correlations between different pairs of variables, viz. polarization index (PI), water level (WL), and fractional water surface area (F WS), were also investigated. At the X band, the water level was better correlated with the PI than with emissivity and other parameters defined in the literature. The correlation was good in cases of slow variation in WL. In cases of sudden variation in the river, the PI followed the variations with some time delay related to the propagation of water within the covered AMSR-E pixel.

L-band emission from a Eucalyptus forest in various soil conditions during the nafe campaign

This paper analyzes the L-band radiometric signatures collected over a Eucalyptus forest with moderate biomass, and compares the results with simulations of a theoretical model and a simple model. Due to the limited biomass and the thin litter layer, some sensitivity to variations of soil moisture is observed. Simulations of the theoretical model are in general agreement with measurements, with an overall rms error of 3.7 K. The greatest discrepancy is observed for measurements collected just after a rainfall. The performance of the simple model is only slightly worse than that of the theoretical model, leading to an overall rms error of about 5K.

Application of passive microwave and optical signatures to monitor submerging of vegetation due to floods

In this article, passive microwave observations in synergy with optical data are exploited to monitor floods and estimate vegetation submerging. The selected site is Sundarban Delta, at the borders between India and Bangladesh. The area is subject to severe monsoon in summer, producing heavy floods and vegetation submerging. Because of their high spatial resolution, Moderate Resolution Imaging Spectroradiometer (MODIS) signatures are used to evaluate the coverage fractions of bare soil, vegetated fields, and permanent water. Multifrequency Advanced Microwave Scanning Radiometer Earth Observing System (AMSR-E) signatures are used to monitor vegetation submerging during monsoon. Results are compared with ground measurements of water level and plant biomass in both agriculture areas and wetlands. Previous studies indicated that, during monsoon, there is a clear effect of brightness temperature decrease and polarization index increase in the C, X and Ka bands over the areas affected by floods. X band data prove to be particularly useful since the sensitivity to flood effects is appreciable and the spatial resolution is better than at C band. In this article, the vegetation submerging effect is estimated with the aid of a radiative transfer model. In the pre-monsoon season, the retrieved value of emerged biomass is close to that of the measured total biomass. During monsoon, it is estimated that up to 3 kg m−2 of vegetation biomass is submerged by flood. For both agricultural fields and wetlands, obtained results are consistent with ground measurements of water level.

Study of multifrequency sensitivity to soil moisture variations in the lower Bermejo basin

Abstract In this paper, a sensitivity analysis to soil moisture variations as a rain effect has been performed at several microwave bands over the lower Bermejo basin, a subtropical area of Argentina mostly spread by moderately dense forests. Parameters such as emissivity and Polarization Index have been considered to carry out the study. In particular, the performance of L-band SMOS measurements has been compared with C and X band AMSR-E one, highlighting the better achievement of the lower frequencies due to the weaker interaction with the vegetation structures. This work intends to give a contribution in the subject of soil moisture sensitivity, which is a preliminary step in the development of retrieval algorithms.

Use of passive microwave signatures to detect and monitor flooding events in Sundarban Delta

In this paper, the potential of using passive microwave observations to detect and monitor flood was explored during the time period 2005–2009. Data from the Advanced Microwave Scanning Radiometer (AMSR-E) was used to evaluate the performance of radiometry in flood monitoring and detection. The current study demonstrates the potential of these observations to monitor the floods in the Sundarban Delta (India / Bangladesh). There is a clear effect expressed by low brightness temperature and high polarization index (i.e. higher soil moistures) over the affected areas after floods or heavy rainfalls. The use of remotely sensed passive microwave observations can improve the forecasting skill of flooding at regional scale.

Refinements and tests of a microwave emission model for forests

This paper shows model simulations of forest emissivity at L-band and at global scale. The electromagnetic model developed at Tor Vergata University has been combined with information available from forest literature. Using allometric equations and auxiliary information, the geometric and dielectric inputs required by the model have been related to global variables available at large scale, such as Leaf Area Index. Simulations indicate that, at L-band, leaves are almost transparent, attenuation is mostly due to branches, and soil contribution can be still appreciable, unless the forest is dense. The model is being refined, to consider seasonal variations of foliage cover, subdivided into arboreous foliage and understory contribution. Parametric simulations, as well as comparisons with experimental data, are shown.