Table 8.
Main inputs of the retrieval algorithms of SMOS, FY3-B and SMAP SSM.
| Foundational model | Zero-order radiative transfer model (i.e.,τ−w model) | ||
|---|---|---|---|
| Algorithm | L-MEB for SMOS | Qp model for FY3-B | SCA-V for SMAP |
| Brightness temperature (TB) | 1.4 GHz with muti-angles (2.5 ± 2.5 to 62.5 ± 2.5°) | 10.65 and 36.5 GHz | Interpolated TB at 1.41GHz using the Backus-Gilbert method |
| Soil roughness model |
QHN model: Q = 0; NH-p = NV-p = −1 (low vegetation); NH-p = 1, NV-p = −1 (V-p), (forest); H: from LUT based on IGBP (0.02 ≤ H ≤ 0.3) |
Qp model: The coefficients of Qp model are estimated using simulated data from AIEM |
QHN model: Q = 0; N = 2; H: from LUT based on IGBP (0.083 ≤ H ≤ 0.160) |
| Vegetation optical depth (τ) | τ is retrieved in L-MEB with soil moisture simultaneously |
τ = b × VWC/cosθ: VWC: from an empirical function of LAI; θ: the incident angle. |
τ = b × VWC: b: from LUT based on IGBP (0 ≤ b ≤ 0.13); VWC: from the nonlinear function of NDVI |
| Soil temperature (Ts) | from ECMWF (Level 1~3) | from an empirical linear function of TB at 36.5 GHz | from the GMAO-GEOS-FP (Level 1: 5~15cm and Level 2: 15–35 cm) |
| Canopy temperature (Tv) | from ECMWF (Level 1: top 0–7 cm) | Tv = Ts | Tv = Ts |
| Single scattering albedo (w) | from LUT based on IGBP (0.06 ≤ w ≤ 0.12) | w = 0 | from LUT based on IGBP (0 ≤ w ≤ 0.07) |
| Soil dielectric model | Mironov et al. | / | Mironov et al. |
*Q, H and N are the parameters of QHN model; H-p: Horizontal-polarization; V-p: Vertical-polarization; IGBP: the International Geosphere-Biosphere Programme; ECMWF: the European Center for Medium range Weather Forecasting; AIEM: the Advanced Integral Equation Method; VWC: the Vegetation Water Content; LAI: the Leaf Area Index; NDVI: the Normalized Difference Vegetation Index; GMAO-GEOS-FP: the Goddard Modelling and Assimilation Office-Goddard Earth Observing System (model)-Forward Processing; LUT: Look-Up-Table.