Search

Accurate measurement of soil moisture (SM) is crucial for understanding crop growing conditions, optimizing irrigation practices, and early detection of drought. Synthetic aperture radar (SAR) has proven effective in SM inversion for agricultural scenarios. Machine learning methods enable large-scale SM mapping by circumventing complex computations and exhibiting high nonlinear fitting capabilities. While previous studies have explored SM retrieval using SAR data and machine learning with dual-polarimetric (DP), compact-polarimetric (CP), or quad-polarimetric (FP) modes, a comprehensive comparative study of these polarization modes in crop scenario is lacking. In this study, we assessed SM inversion using three SAR polarimetric modes (DP, CP, and FP) in C-band across three crop types (wheat, corn, and soybean) using multi-year RADARSAT-2 data. Various polarimetric backscattering variables, polarimetric decomposition parameters, and vegetation indices under different polarimetric modes were extracted to construct the corresponding SAR feature sets. Four machine learning algorithms, including Bagged Decision Tree (BAGTREE), Random Forest Regression (RFR), Extreme Gradient Boosting (XGB), and Gaussian Process Regression (GPR), were used. Additionally, forward feature selection (FFS) procedure was employed to reduce redundant features and enhance accuracy. Results indicate that FP mode consistently demonstrated superior performance in SM retrieval, with CP mode slightly trailing behind, and DP mode yielding the least favourable outcomes. The FFS method consistently enhanced SM retrieval accuracy. Among the machine methods, RFR and GPR exhibited superior performance across all three crop scenarios and three polarimetric modes. Specifically, RFR achieved the best accuracy in the corn and soybean scenarios, with root mean square errors (RMSE) of 4.46 vol.% and 6.49 vol.%, respectively, while GPR excelled in the wheat scenario, with a RMSE of 4.29 vol.%. FFS outputs highlig

In this study, a general three-component polarimetric SAR interferometry (PolInSAR) target decomposition framework is proposed by modifying the existing generalized surface, double-bounce, and volume models. The resulting general models are then compared to the original Freeman-Durden modeling strategy. Three types of generalized volume scattering models (generalized volume scattering model (GVSM), simplified Neumann volume scattering model (SNVSM) and simplified adaptive volume scattering model (SAVSM)) were employed. Simulated L-band PolInSAR data over deciduous and pine forest stands generated by PolSARpro and DLR P-band airborne PolInSAR data over a tropical forest area from the AfriSAR 2016 campaign were used for performance analysis. A qualitative comparison of the decomposition results shows that the three generalized volume scattering models generally deviate from the Freeman-Durden model, showing that the GVSM and SNVSM models have very similar results. In the case of airborne data over tropical forests, a tomographic synthetic aperture radar (TomoSAR) profile was also computed and used as a benchmark for comparison with the phase-center profiles of all four volume-scattering components. Not only do the GVSM and SNVSM models exhibit similar results between them (as with simulated data), but also a better match with the HV TomoSAR profile.

Monitoring crop phenology is essential for managing field disasters, protecting the environment, and making decisions about agricultural productivity. Because of its high timeliness, high resolution, great penetration, and sensitivity to specific structural elements, synthetic aperture radar (SAR) is a valuable technique for crop phenology estimation. Particle filtering (PF) belongs to the family of dynamical approach and has the ability to predict crop phenology with SAR data in real time. The observation equation is a key factor affecting the accuracy of particle filtering estimation and depends on fitting. Compared to the common polynomial fitting (POLY), machine learning methods can automatically learn features and handle complex data structures, offering greater flexibility and generalization capabilities. Therefore, incorporating two ensemble learning algorithms consisting of support vector machine regression (SVR), random forest regression (RFR), respectively, we proposed two machine learning-aided particle filtering approaches (PF-SVR, PF-RFR) to estimate crop phenology. One year of time-series Sentinel-1 GRD SAR data in 2017 covering rice fields in Sevilla region in Spain was used for establishing the observation and prediction equations, and the other year of data in 2018 was used for validating the prediction accuracy of PF methods. Four polarization features (VV, VH, VH/VV and Radar Vegetation Index (RVI)) were exploited as the observations in modeling. Experimental results reveals that the machine learning-aided methods are superior than the PF-POLY method. The PF-SVR exhibited better performance than the PF-RFR and PF-POLY methods. The optimal outcome from PF-SVR yielded a root-mean-square error (RMSE) of 7.79, compared to 7.94 for PF-RFR and 9.1 for PF-POLY. Moreover, the results suggest that the RVI is generally more sensitive than other features to crop phenology and the performance of polarization features presented consistent among all methods, i.

Forest canopy height estimation is essential for forest management and biomass estimation. In this study, we aimed to evaluate the capacity of TanDEM-X interferometric synthetic aperture radar (InSAR) data to estimate canopy height with the assistance of an external digital terrain model (DTM). A ground-to-volume ratio estimation model was proposed so that the canopy height could be precisely estimated from the random-volume-over-ground (RVoG) model. We also refined the RVoG inversion process with the relationship between the estimated penetration depth (PD) and the phase center height (PCH). The proposed method was tested by TanDEM-X InSAR data acquired over relatively homogenous coniferous forests (Teruel test site) and coniferous as well as broadleaved forests (La Rioja test site) in Spain. Comparing the TanDEM-X-derived height with the LiDAR-derived height at plots of size 50 m × 50 m, the root-mean-square error (RMSE) was 1.71 m (R2 = 0.88) in coniferous forests of Teruel and 1.97 m (R2 = 0.90) in La Rioja. To demonstrate the advantage of the proposed method, existing methods based on ignoring ground scattering contribution, fixing extinction, and assisting with simulated spaceborne LiDAR data were compared. The impacts of penetration and terrain slope on the RVoG inversion were also evaluated. The results show that when a DTM is available, the proposed method has the optimal performance on forest height estimation.