| Paper: | TH-A2.28 |
| Session: | Applications of Radiometry II |
| Time: | Thursday, March 29, 09:00 - 10:20 |
| Presentation: |
Poster
|
| Topic: |
Theory, physical principles and electromagnetic models: |
| Title: |
A soil moisture retrieval scheme based on a UAV distributed architecture and bistatic scattering models |
| Authors: |
Esteban Roitberg; Institute of Astronomy and Space Physics, Quantitative Remote Sensing Group - CONICET | | |
| | Mariano Franco; Institute of Astronomy and Space Physics, Quantitative Remote Sensing Group - CONICET | | |
| | Francisco Grings; Institute of Astronomy and Space Physics, Quantitative Remote Sensing Group - CONICET | | |
| | Pablo Perna; Institute of Astronomy and Space Physics, Quantitative Remote Sensing Group - CONICET | | |
| Abstract: |
Soil moisture is an essential variable for agrometeorological applications, since it is the most common limit to crops’ yield. Indeed, the availability of water at plot scale in key crop stages limits overall productivity and is therefore a major concern for farmers. Active microwave remote sensing observations can provide valuable mapping information of soil moisture, due to the well known relation between backscattering and soil dielectric constant. This is particularly useful for UAV-based radar platforms, which are currently in use to estimate soil moisture at plot scale on demand. UAV-based instruments offer high versatility and flexibility, as compared to airborne systems or satellites, and can operate rapidly without planned scheduling. Additionally, they can fly at low altitudes and slowly, with the ability of acquiring spatial and temporal high resolution data. However, the backscattered signal (the only one available for single platform systems) is also influenced by surface roughness, vegetation, and topographic effects. This is why the retrieval of soil moisture using backscattering-only measurements is an ill-posed problem.
In this context, the availability of UAVs platforms allow to fully exploit the distributed architecture paradigm. In particular, the radar transmitter and receiver can be mounted on different platforms, in order to measure the reflected signal in any practical geometry. To fully exploit this setup for extracting relevant soil information, theoretical models that compute the bistatic scattering function are crucial. Particularly, for random rough surfaces, the scattered field can be computed through approximated or perturbative methods that have a validity range limiting their application. Two of most commonly used are the small perturbation method (SPM) and tangent plane -or Kirchhoff- approximation (KA). These two theoretical models allow to compute the bistatic scattered wave by the surface in terms of its geometrical and dielectric parameters (RMS height, correlation length and dielectric constant). The main advantage of using these models in the framework of UAVs platforms is that both models are suitable to compute the scattering function in real time for a soil moisture retrieval scheme. One key result of simulations based on these models is that the soil cross-polarized coefficient is maximum in directions considerably outside the incidence plane and its magnitude is similar to the copol cases in forward and backward scattering condition.
In this paper we evaluate the feasibility of a “distributed architecture” bistatic radar for soil moisture retrieval. The setup proposed consists in two UAVs, one with a radar emitter (H-V) and the other one with a receiver (H-V). This allows to potentially measure the full scattering matrix in different geometrical configurations. In particular, it is possible to measure HV outside the incidence plane. Moreover, we present a retrieval algorithm and its corresponding flight plan for the acquisition of the elements of the scattering matrix in key places of interest. Using an OSSE, we show that the proposed observational setup is able to estimate the dielectric constant and surface roughness of the bare soil with reasonable errors. Results are compared with the benchmark case (backscattering only), and the trade off of the proposed architecture is evaluated. |
