Synthetic Aperture Radar (SAR) Compact Polarimetry for Soil Moisture Retrieval

Abstract

Soil moisture is a factor for risk analysis in the agricultural sector, yet access to temporally and spatially detailed data is challenging for much of the world’s agricultural extend. Significant effort has been focused on developing methodologies to estimate soil moisture from microwave satellite sensors. Canada’s RADARSAT Constellation Mission (RCM) is capable of acquiring imagery in a number of modes with a Compact Polarimetry (CP) configuration at different spatial resolutions (1 to 100 m). RCM offers greater polarization diversity, wide swaths and improved temporal frequency (4-day exact revisit time); all important considerations for large area monitoring of agricultural resources. The major goal of this study was to examine whether CP could accurately estimate surface soil moisture over bare fields. A methodology was developed using the calibrated Integral Equation Model (IEM) multi-polarization inversion approach. RADARSAT-2 data was acquired between 2012 and 2017 over a test site in eastern Canada. CP backscatter for two RCM modes (medium resolution 30 m and 50 m (MR30 and MR50)) was simulated using 63 RADARSAT-2 fully polarimetric images. A simple transfer function was developed between RH (right circular-horizontal) and HH (horizontal-horizontal) intensity, as well as RV (right circular-vertical) and VV (vertical-vertical). These HH- and VV-like intensities were then used in the multi-polarization inversion scheme to retrieve soil moisture. CP soil moisture retrievals were validated against soil moisture measurements from a long term in-situ network instrumented with five soil moisture stations. Retrieved and measured soil moisture were well correlated (R > 0.70) with an unbiased root mean square error (ubRMSE) less than 0.06 m3/m3. Overall, the developed method clearly captured the dry down and wetting trends observed through the five years study period. However, results demonstrated that the inversion method introduced a consistent bias (~0.10 m3/m3). Comparison of CP soil moisture estimates to those from the Soil Moisture Active Passive (SMAP) passive microwave satellite confirmed this bias. This study demonstrates the potential of C-band CP data to deliver accurate soil moisture products over wide swaths for regional and national soil moisture monitoring.

Combining Machine Learning and Compact Polarimetry for Estimating Soil Moisture from C-Band SAR Data

“…In the case of Casselman, two RS2 FQW images were acquired in November 2014 and June 2015, respectively (Table 1). Both images were acquired over bare soil conditions, since the first image (November 2014) was acquired after the growing season and the second image (June 2015) was acquired in the spring when crops were not present [36]. The bare soil conditions in June 2015 was further confirmed by considering the Normalized Difference Vegetation Index (NDVI) derived from the closest Landsat-8 L1TP image over Casselman in almost cloud free conditions, which was collected on 29 May 2015.…”

“…The location of the two test areas is shown in Figure 1. The RS2 polarimetric acquisitions (24 + 2) were processed by using a CP SAR simulator developed for the RADARSAT Constellation Mission (RCM) at the Canada Centre for Mapping and Earth Observation [36,37]. The simulator converts the RADARSAT-2 16-bit complex products to a 32-bit float complex according to the product documentation [38] by applying the sigma-nought calibration coefficients provided in the data files.…”

Sequential 90° Rotation of Dual-Polarized Antenna Elements in Linear Phased Arrays with Improved Cross-Polarization Level for Airborne Synthetic Aperture Radar Applications

“…Due to the penetrating nature of electromagnetic waves at higher wavelengths and by means of interferometry, SAR becomes an interesting way to monitor forest density and estimate forest height by using tomography [8], as well as to perform geology and glaciology studies for the measurement of snow, ice layer thicknesses and soil moisture. Furthermore, SAR polarimetry allows the use of polarization diversity in order to identify different scattering mechanisms, enhancing this way the retrieval of information [9].…”

RADOLAN_API: An Hourly Soil Moisture Data Set Based on Weather Radar, Soil Properties and Reanalysis Temperature Data

“…However, given the limitations regarding cloud cover a continuous stream of satellite sensed soil moisture is only possible with active and passive microwave sensors like the Advanced Scatterometer (ASCAT) [24] or the Soil Moisture Active Passive (SMAP) Mission [25,26]. Various algorithms for these platforms have been developed [18,[27][28][29][30][31][32]. The spatial resolution of such data sets and retrieval algorithms usually is rather coarse with tens of kilometers [26].…”