The Remote Sensing Hub (RSHub) is a shared computing platform intended for education and research to model microwave scattering properties including but not limited to brightness temperatures and backscatters via microwave scattering models. Currently, it supports scenarios of bare soil, snow-covered soil, and vegetation-covered soil. The Numerical Maxwell Model in 3D (NMM3D) - based full wave soil models, Vector Radiative Transfer (VRT) - based vegetated land surface scattering models, and Dense Media Radiative Transfer (DMRT) - based snow models have been integrated into RSHub. It is a useful tool to study the relationship between microwave signals and scattering properties under different sensor configurations such as frequency, incident angles, and polarizations under varying scenarios for developing mission-specific retrieval algorithms. A user-friendly Python toolbox is available to connect to the RSHub to run models and retrieve results with a unified calling interface, alleviating the complexity of setting up computing environments and computing resources locally. This tutorial presents the interaction structure, the supported scenarios, and the integrated models of RSHub. A demo application of RSHub is presented to validate the modeled brightness temperature against measurements from the soil moisture active passive validation experiment 2012 (SMAPVEX12) under a forested scenario.

In microwave remote sensing of vegetated land surfaces, there is volume scattering from the vegetation and rough-surface scattering of the interface between the vegetation and the ground. However, the coupling of volume scattering and surface scattering and its impact on microwave emission have not been fully studied, particularly when multiple scattering effects are to be considered, as in forested scenarios. Earlier treatment with the Kirchhoff approximation (KA) neglects the incoherent bistatic scattering components from the rough soil interface and thus overestimates the transmitted upward emission from the soil half-space and underestimates the reflected upward emission from the downward-going intensity streams arising from the vegetation layer in nonspecular directions. To address this issue, bistatic scattering coefficients are introduced into the boundary conditions, and an interpolation-accelerated numerical iterative method is employed to solve the passive radiative transfer (RT) equation to rigorously couple the volume–surface scattering interactions at the vegetation/soil interface. In this article, the rough-surface bistatic scattering is modeled using the advanced integral equation model (AIEM). The volume scattering model is derived from an RT-based multiple scattering model that accounts for the vertical profile of the vegetation structure. The SMAPVEX12 forest dataset is utilized to validate the proposed model, encompassing L-band radiometric brightness temperature (TB) observations corresponding to extensive variations in soil moisture. Comparisons are made between the TBs simulated by the model with and without considering the incoherent bistatic scattering coefficients, under diverse vegetation water contents (VWCs) and varying soil roughness conditions. Notable features are observed in the angular pattern of the vegetated surface emission by fully accounting for the multiple scattering effects and the incoherent bistatic scattering effects of the rough soil. These features become more evident at smaller observation angles, lower VWCs, and greater soil roughness. The findings reveal that the model incorporating the bistatic scattering coefficient of rough surfaces exhibits improved agreement with the measured TBs. Furthermore, the proposed model is parameterized by matching the high-order solutions to the RT equation to the widely adopted albedo–tau formalism, that is, the zero-order solution to the passive RT equation with a flat lower boundary. The resulting equivalent optical thickness and the equivalent scattering albedo incorporate the multiple scattering effects within the vegetation layer, and they are solely associated with the geometries and the electromagnetic properties of the vegetation layer. Additionally, the effective reflectivity of the soil is proposed to characterize the scattering properties of rough surfaces and the scattering coupling between the rough surface and the vegetation layer. The new model developments will enhance the prediction and interpretation of vegetated land surface emission characteristics and thus improve the remote sensing of the vegetation and the underlying soil parameters.

In this paper, a novel framework is proposed which combines physical scattering models with artificial neural networks (ANN) to solve electromagnetic scattering problems of random media through a volume integral equation formulation. The framework is applied to a snow scattering problem where snow is represented by a bicontinuous random medium. A neural network is constructed linking the random media structure to the induced dipole moments on the media. The volume integral equation (VIE) serves as a natural physical constraint on the network input-out relations and is used to guide the training of the network. A discrete dipole approximation (DDA) strategy is adopted to convert the VIE into matrix equations which also defines the loss function of the surrogate neural network. For addressing deterministic scattering problems, this represents a viable alternative to traditional iterative algorithms, providing comparable accuracy at the expense of reduced efficiency. In solving statistical scattering problems, neural networks with physics-informed loss function achieve accuracy comparable to that of data-driven models while significantly reducing the dependency on extensive precomputed training datasets. The physics-based loss function also allows the network to self-diagnose the prediction accuracy in real operations. This work demonstrates a novel strategy to effectively merge physical equations with artificial neural networks, and the idea can be inspiring to many relevant fields, especially when randomness effects are exhibited through a complicated nonlinear system.