In this work, we proposed a reappraisal of strong fluctuation theory (SFT) aimed at improving its capability in simulating snow backscattering. As conceived in the original form, SFT considers the air-snow and snow-soil interfaces as flat surfaces. This study aims at accounting for the roughness of the snow-soil interface by coupling SFT with suitable rough soil surface models, namely, Oh model and integral equation model (IEM), with the aim of increasing accuracy and reliability of SFT simulations. The accuracy of the reappraised SFT was evaluated in comparison with SAR data collected by Sentinel-1 (S-1) and COSMO-SkyMed (CSK) over two test areas located in the western part of Italian Alps, for which in situ measurements of the main soil and snow parameters were available. The reappraised SFT was able to significantly improve the simulation accuracy at both frequencies with respect to the original implementation. In the better case, we obtained$R^{2} =0.57$, RMSE =2.9 dB, and bias$= -0.17$dB. A comparison with dense media radiative transfer (DMRT) theory in single-layer configuration demonstrated that the reappraised SFT can obtain comparable results with a significantly smaller computational cost.

In this paper the potential of integrating optical and microwave data to monitoring vegetation features has been exploited by using experimental data and models. The general idea was to cope the high sensitivity of radar data to water content of vegetation with the high sensitivity of optical data to pigments, thus producing more in-depth information on vegetation status. Two sorghum fields located close to Florence was taken under observation during summers 2022 and 2023, by gathering soil and vegetation parameters and collecting Sentinel-1 and Sentinel-2 images. Backscattering coefficient and some optical indices have been experimentally related to soil and vegetation water content and plant water status. The use of a simple e.m. model allowed estimating the plant water content in the canopy. The obtained results confirmed the validity of the followed approach, although further investigation is needed.

In this study, the retrieval of snow Liquid Water Content (LWC, %) from C- and X- band SAR data was based on Artificial Neural Network (ANN) and Random Forest (RF). Two approaches were explored for generating a sufficient amount of data to train and test the ANN and RF algorithms: the first strategy was defined as “model-driven”. The second one was defined as “data-driven”. The validation results showed that RF performs better than ANN in terms of correlation coefficient R, regardless of the selected approach ("model driven" RANN= 0.60, RRF= 0.68; “data-driven” RANN= 0.50, RRF= 0.88 at X-band). Moreover, the RF implementation trained with the “data-driven” approach outperformed the “model-driven” approach in terms of correlation coefficient R (RRF= 0.88 at X-band).

This study aims at improving the spatial resolution of snow depth (SD) products derived from microwave satellite radiometers by proposing a disaggregation method based on X-band SAR data. The method has been developed and tested in the Western part of Italian Alps, by involving Cosmo SkyMed (CSK) and AMSR-2 data. Machine learning methods play a twofold role in the proposed active/passive (A/P) implementation: the AMSR-2 data disaggregation process is indeed based on Artificial Neural Networks (ANN), while the SD retrieval using the disaggregated data is based on ANN and Random Forest (RF) algorithms. To assess the effectiveness of the proposed A/P technique, the SD retrievals have been compared with those obtained by estimating SD directly from CSK data. Taking advantage of the multifrequency information, the retrievals based on A/P method clearly outperformed those based on CSK data only: correlation increased from R=0.77 to R= 0.85 for the ANN based retrievals and from 0.76 to 0.86 for the RF based retrievals. The corresponding RMSE decreases from 34 cm to 28 cm and from 34 cm to 27 cm for ANN and RF, respectively, in a SD range between 0 and ≃ 220 cm.

This study aims at improving the accuracy of the Strong Fluctuation Theory (SFT) in simulating the backscattering from Alpine snowpacks, by simulating the roughness effect of the snow-soil interface through suitable models, as the Oh model and the Advanced Integral Equation Model (AIEM). As conceived in the original form indeed, SFT considers the air-snow and snow-soil interfaces as flat surfaces: such approximation can lead to inaccurate results under some observed conditions. The reappraised SFT was validated against Dense Media Radiative Transfer (DMRT) model simulations and experimental data available from Sentinel-1 (S-1) C-band and COSMO-SkyMed (CSK) X-band SAR in two alpine test sites located in the Northern Italy. The inclusion of rough soil contribution was found effective in improving significantly the SFT simulation in dry and wet snow conditions, with a significant improvement of correlation with SAR data: as an example, R2increased from 0.05 to 0.57 in the comparison with CSK. The comparison with DMRT pointed out a very good agreement between the two models, (R2=0.88 at C-band and 0.91 at X-band) with the not negligible advantage of an extremely reduced computational cost of the reappraised SFT with respect to DMRT.

This research aims to exploit the potentialities of multi-mission SAR data at X-, C- and L-band for the monitoring of snowpack and alpine soils. The snow parameters as snow water equivalent, snow liquid water content and snow metamorphism have been monitored and different methods are proposed for their retrieval. In order to gather consistent datasets, experimental activities have been conducted in two selected sites in Northern Italy, which are covered by alpine snow during winter and spring periods and are in some cases characterized by the presence of permafrost. Microwave responses of snow and soil have been then simulated by using electromagnetic (i.e., AIEM, Oh, SFT and DMRT-QCA), and physical models (SNOWPACK). Finally, machine learning approaches, as Artificial Neural Networks and Random Forest, were implemented for retrieving snow parameters; whereas interferometric techniques were used in case of snow and soil displacement as rock glaciers. Preliminary and consistent results have been obtained in terms of estimate of snow parameters and soil displacement. This multi-frequency/multi-mission approach enhances the ability of SAR sensors to monitor and analyze snow dynamics, contributing to improved decision-making in various domains.

In this study, a method for mapping the crop biomass, expressed as Plant Water Content (PWC in kg/m2), at high resolution is proposed. The method is based on SAR data at C and X bands and machine learning algorithms, and it is composed of some steps, including crop classification, soil moisture (SMC) retrieval and finally PWC retrieval. It has been developed and validated in an agricultural area located in Tuscany (Central Italy), for which timeseries of Sentinel-1 and COSMO-SkyMed images were available, along with in situ measurements of the main soil and vegetation parameters.The retrieval, so far limited to the wheat crops, resulted in correlation coefficient R=0.92 and RMSE=0.5 (kg/m2) between estimated and target PWC, by confirming the feasibility of using SAR for monitoring vegetation biomass at high resolution.

This paper deals with the analysis of the backscattering sensitivity at C and X bands to the agricultural crop characteristics and the implementation of a method for estimating crop biomass. The study areas were located in Tuscany (Central Italy) close to Florence. Series of Sentinel-1 and COSMO-SkyMed images have been collected for several years. An accurate crop classification method was first realized in order to separate crops characterized by different scattering behaviors, namely broad- and narrow-leaf crops. The backscattering trends have been simulated by using electromagnetic models based on radiative transfer theory. Algorithms based on Neural Network approaches have been implemented for estimating the crop biomass by using multi-frequency and multi-polarization SAR data at C and Xband.

The research results described in this paper have been obtained in the framework of the 2019–2022 ALGORITMI project between the Italian Space Agency (ASI) and the Institute of Applied Physics of the National Research Council (CNR-IFAC). The focus of the research was the development of innovative algorithms for the estimation of geophysical parameters of soil, snow, and vegetation with the aim of monitoring soil, snow cover and agricultural crop conditions. The estimation of soil moisture, vegetation biomass, snow water equivalent, and crop classification was improved by using retrieval algorithms based on machine- learning approaches and temporal series of SAR images from COSMO-SkyMed (CSK) and Sentinel-1 (S-1) missions, along with optical images from Sentinel-2. This paper provides an overview of the most recent and valuable results obtained during the project. In particular, the validation of soil moisture provided R=0.89 and RMSE=0.025 m3/m3by integrating data from S-1 and CSK and that one of snow water equivalent gave R=0.85 with RMSE=86.24 mm (CSK HIMAGE) and R=0.86 with RMSE=71.59 mm (CSK PP). Early mapping results showed an almost monotonic progression in overall accuracy over time higher than 90% by increasing the available images.

In this paper, an investigation on the agricultural management in semi-arid Mediterranean regions is presented. The selected test areas are located in Tunisia, near the Kairouan town. The agricultural fields are mainly cultivated with olive trees together with cereals, fruit trees and vegetables. The possibility to monitor this area by means of COSMO-SkyMed (CSK) data, thanks to the ASI Open Call initiative, is an added value to retrieve information concerning the temporal evolution of crop conditions and the use of water in semi-arid regions. The CSK images have been acquired in the period 2018–2019 and the spring 2021. The objectives of this research concern the use of CSK data to evaluate the correct growth of agricultural crop. The preliminary analysis shows that X -band backscatter is able to follow the seasonal moisture conditions and to identify different types of crops.

In this study, a disaggregation technique based on machine learning is proposed. The technique combines Sentinel 1 and AMSR2 data with the aim of enhancing the spatial resolution of the vegetation biomass, expressed herein as Plant Water Content (PWC), and Soil Moisture (SM) products generated from AMSR2 by the HydroAlgo algorithm developed at IFAC. Validation is still in progress; however, the results obtained so far demonstrated the effectiveness of the proposed disaggregation in mapping both PWC and SM at 100m resolution, thus overcoming the problem of coarse spatial resolution that hampers the potential of satellite microwave radiometers as the AMSR2 for operational applications in small scale basins.

This study aims at relating the stickiness parameter (τ) of the Dense Media Radiative Transfer theory integrated with Sticky Hard Sphere (SHS) model (DMRT-QMS), to the physical parameters of the layered snowpack. A relationship has been derived to express τ, which modulates the attractive contact force between ice spheres, as a function of ice volume fraction (ϕ) and coordination number (nc). Since τ is not a measurable parameter, this is a step forward with respect to what is commonly made in literature, where τ is assumed as an arbitrary parameter, generally ranging between 0.1 and 0.3, to fit simulated backscattering data with those measured. As a first validation, DMRT-QMS was integrated with SNOWPACK model to simulate backscattering at X band (9.6 GHz) driven by nivo-meteorological data acquired on a test area located in Monti Alti di Ornella, Italy. The simulations were compared with Synthetic Aperture Radar COSMO-SkyMed (CSK) satellite observations. The results show a significant agreement (R2=0.68), although for a limited dataset of eight points in a unique winter season.

In this paper the integration between the Dense Media Radiative Transfer (DMRT-QMS) model and the SNOWPACK model was investigated in order to simulate snow parameters and the backscattering at X band (9.6 GHz) from nivo-meteorological data. The role of the stickiness parameter ($\tau$) in DMRT-QMS was analyzed by using experimental data of backscattering collected from COSMO-SkyMed (CSK) and snow data generated by SNOWPACK. The relationships between$\tau$and both ice volume fraction ($\phi$) and coordination number ($n_{c}$) were assessed. The DMRT and SNOWPACK simulations were compared with CSK backscattering measurements showing a significant agreement, although for a limited dataset.

In this paper, an algorithm based on Convolutional Neural Networks (CNNs) was developed to correctly classify an agricultural area in central Italy, by using SAR images. This preliminary step is vital for mastering the different influence of crop types in SAR data before the implementation of algorithms devoted to estimate of vegetation biomass. In situ data collected on the test site were used for validating the CNN algorithm-based classification. After the agricultural species recognition, a sensitivity analysis between C-band Sentinel-1 and X-band COSMO-SkyMed backscatter coefficients and crop biomass was carried out, laying the foundation for the implementation of algorithms able to estimate the biomass of different crop types.