| Paper: | WE-P1.2 |
| Session: | Biosphere Applications of Radiometry |
| Time: | Wednesday, March 28, 13:40 - 14:00 |
| Presentation: |
Oral
|
| Topic: |
Soil moisture, soil state and vegetation: |
| Title: |
SMAP Soil Moisture and Vegetation Attenuation and Scattering Retrievals Including Higher Order Soil-Canopy Interactions |
| Authors: |
Andrew Feldman; Massachusetts Institute of Technology | | |
| | Ruzbeh Akbar; Massachusetts Institute of Technology | | |
| | Dara Entekhabi; Massachusetts Institute of Technology | | |
| Abstract: |
NASA’s Soil Moisture Active Passive (SMAP) and ESA’s Soil Moisture Ocean Salinity (SMOS) projects rely on a zeroth order radiative transfer model, known as the tau-omega model, to retrieve soil moisture from microwave brightness temperature observations. The model assumes the vegetation layer is a weakly scattering medium which is not the case at L-Band in forests, or 30% of land cover, where canopy constituents exceed the microwave wavelength. Previously developed higher order radiative transfer models have proven effective in characterizing emission from vegetated surfaces, but are difficult to invert in retrieval algorithms and require extensive in-situ measurements and characterization of canopy layer physical properties. We propose a first order radiative transfer model that approximately characterizes eight first order emission pathways using rough surface reflectivity, vegetation optical depth (VOD), and scattering albedo (ω) terms. It introduces a first order scattering coefficient (ω1) which represents scattering from a woody biomass medium. This first order radiative transfer model is determined to be more sensitive to surface emission than the tau-omega model. Taking advantage of this simple model formulation, we replaced the tau-omega model with this first order radiative transfer model in the recently developed Multi-Temporal Dual Channel Algorithm (MT-DCA) which retrieves soil moisture and VOD along with a temporally constant ω using brightness temperature observations. We further developed this algorithm with the goal of additionally retrieving temporally constant ω1 in order to characterize scattering from woody biomass based solely on SMAP observations without ancillary vegetation and land use information. This first order retrieval algorithm was run over Africa using SMAP L1C brightness temperature inputs from April 2015 to March 2016 at both 36 and 9 km resolutions. The algorithm generally finds non-zero ω1 over regions of woody biomass and zero ω1 over bare to lightly vegetated surfaces as expected. Based on this result, we assert that the first order retrieval algorithm is able to detect scattering from woody biomass using only SMAP brightness temperature observations. It is also possible to retrieve soil moisture and VOD while accounting for higher order scattering especially in vegetated regions. |
