soil moisture retrieval based on a coherent analysis of ... · prospect for small satellites...

Soil moisture retrieval based on a coherent analysis of Illuminators Of Opportunity: a prospect for small satellites applications

Francisco Grings - IAFE Remote Sensing Group

Motivation I

• Soil moisture (SM) is a key biophysical variable that controls the partition of both mass (rain, runoff) and energy (evapotranspiration) fluxes on the soil and therefore is important for the monitoring of many hydro-geochemical cycles.

• From the agricultural point of view, it effectively limits crop productivity. Therefore, the generation of regional-scale soil moisture maps is a critical task that can only be accomplished systematically using remote sensing techniques.

1st IAA Latin American Symposium on Small Satellites - F. Grings - IAFE 2

Motivation II• Drought monitoring

The Pampas in Argentina is one of world's most important grain suppliers. The rapid expansion of crops over pastures, combined with climate change and an extremely flat topography, make this region vulnerable to unpredictable hydrological shifts. Extreme droughts are typically observed.

← drought in Pampas plain as seen by SMOS, ASCAT and GLDAS.

31st IAA Latin American Symposium on Small Satellites - F. Grings - IAFE

Traditional approaches to estimate SM from RS data

• Soil moisture = Microwave

Traditionally, surface soil moisture is estimated using the signal reflected from the soil in the microwave regime, using the well known relation between soil backscattering and dielectric constant and dielectric constant and soil moisture.


↑ Modelled relation between p =HH/VV (in backscattering) and soil dielectric constant and normalized roughness.

← empirical relation between dielectric constant and soil moisture

New approach:SM estimation from Illuminators Of Opportunity (IOO)

lluminators Of Opportunity (IOO) present several operative advantages to estimate soil moisture in the framework of small satellites.

1. IOO bistatic radar configurations do not require a dedicated transmitter, which significantly reduces implementation costs.

2. At L Band, there is evidence that soil forward scattering is more sensitive to soil moisture than soil backscattering (e.g. Guerriero et. al.).

Solution: use GNSS (GPS) as an IOO for soil moisture retrieval


Typical SM retrieval techniques based on GNSS

• Typical SM retrieval techniques based on IOO relies on the measurement of the soil forward scattering coefficient, which is related to soil moisture through the dielectric constant.

• Main benefit: the ratio Er/Et is a direct proxy of soil moisture

• Drawback: good antenna isolation is required to separate direct and reflected components (reflected component is usually -20 to -40 dB below direct component)

←conceptual model for the CYGNSS mission, including surface coverage


An alternative: SM estimation from IOO using the Interference Pattern Technique (ITP)

• As an alternative, for IOOs characterized by sufficiently long pulses, the Interference Pattern Technique (IPT) can be used.

• This technique consists on measuring the angular power fluctuations of the interference between the direct and the reflected electric fields as the platform or the IOO moves. This happens because the two signals (direct and reflected) arrive at the antenna almost simultaneously and therefore are coherently added.


SM estimation from IOO using the Interference Pattern Technique (ITP) - Fresnel

• Since most IOO emits at circular polarization, the technique is based on measuring the vertically polarized component of the received signal.

• This “vertically polarized interference pattern” is analyzed in order to find a minimum (“notch”), which according to theory should correspond to the Brewster angle, which is itself related to surface dielectric constant and soil moisture. However, the relation between the angle in which the notch is found and the presence of the notch itself is based on a simplified analysis of soil specular scattering, by assuming a plane interface (Fresnel! no soil roughness!).

← received power as function of time (incidence angle). Adapted from (Rodriguez alvarez et. al.)


SM estimation from IOO using the Interference Pattern Technique (ITP) - SPM 2nd order

• What is needed: a more advanced scattering model

• We implemented the 2nd order Small Perturbation Model (SPM) for the system geometry, to simulate vertically polarized interference pattern expected in the antenna as a function of both geometric and soil properties (dielectric constant and roughness).

← A perturbation series is made in both, the amplitudes of surface fields and the height surface. For L Band, valid when soils RMS height s < 1.2 cm


SM estimation from IOO using the Interference Pattern Technique (ITP) - SPM 2nd order

• Therefore, we obtained a model which computes the sum of the field scattered by the soil and the direct component in the antenna as a function of surface characteristics (dielectric constant and roughness), assuming incident circular polarization and received vertical polarization

← The received power in the antenna is the coherent sum of both fields


E = |Er+Ed|

SM estimation from IOO using ITP - SPM-2 sensitivity analysis for epsilon

• The simulations show the typical pattern predicted in the ITP, which is a function of the wavelength and antenna height.

• As expected, the SPM predicts the notch, and its monotonic behavior as a function of the soil dielectric constant for a typical value of soil RMS height (s = 1 cm, no till condition).


SM estimation from IOO using ITP - SPM-2 sensitivity analysis for roughness

• However the notch amplitude depends strongly on soil roughness, and it disappears for rough soils.

• Therefore, for L band the notch will be seen only only for moderate soil roughness (bare soil).


SM estimation from IOO using ITP and SPM-2: experimental results

• Experimental results obtained using an antenna over a tripod and GPS satellites shows that SPM theoretical predictions present a reasonable agreement with the observations (validation still pending).

←Prototype antenna and typical measurements


SM estimation from IOO using ITP and SPM-2: experimental results

← In this example, the notch is found at 12,1 deg by the retrieval algorithm, that corresponds to eps = 21.6 and s = 0.9 cm.

The mean value of eps = 21.2 ± 3 (measured in situ using a Stevens Hydra Probe II)

Results are promising for bare soils


SM estimation from IOO using ITP and SPM-2: limitations

• It is important to note that this technique cannot be used with GNSS (GPS) systems for orbital systems, since pulse duration is too short (for standard GPS, maximun platform height = 300 m).

• However, other IOOs (like geostationary systems), could be used instead to develop an application based on this techniques.


Summary

Illuminators Of Opportunity (IOO) present several operative advantages to estimate soil moisture in the framework of small satellites.

1. IOO bistatic radar configurations do not require a dedicated transmitter, which significantly reduces implementation costs.

2. At L Band, soil forward scattering is more sensitive to soil moisture than soil backscattering.

3. Standard techniques which rely on measuring soil forward scattering require good antenna isolation to separate direct and reflected components.

4. The Interference Pattern Technique (ITP) uses a single antenna, which is supposed to pick up the two fields at once (less requirements on hardware). Antenna calibration is also not a priority, since the algorithm looks for a minimum in the pattern (notch). When analyzed with sophisticated models, promising experimental results are being obtained.

5. IOO with longer pulses are required for orbital applications.


Thanks!

F. Grings - IAFE 17

soil moisture retrieval based on a coherent analysis of ... · prospect for small satellites...

Documents