H31O-08:
Satellite Retrievals of Vegetation Optical Depth Using Time-Series of Dual-Polarized and Single Look-Angle Global Microwave Observations

Wednesday, 17 December 2014: 9:45 AM
Maria Piles1,2, Alexandra G Konings3, Kaighin A Mccoll3, Steven Chan4 and Dara Entekhabi3, (1)Universitat Politecnica de Catalunya, Barcelona, Spain, (2)SMOS Barcelona Expert Center, Barcelona, Spain, (3)Massachusetts Institute of Technology, Civil and Environmental Engineering, Cambridge, MA, United States, (4)NASA Jet Propulsion Laboratory, Pasadena, CA, United States
Abstract:
Our ability to close the Earth’s carbon budget and predict feedbacks in a warming climate depends critically on knowing where, when and how carbon dioxide is exchanged. Vegetation biomass is an important carbon sink that varies significantly over annual and inter-annual timescales. At global scales, the only feasible approach for monitoring vegetation biomass is satellite remote sensing. In this regard, existing passive microwave missions have the potential of estimating Vegetation Optical Depth (VOD), an indicator of total aboveground vegetation water content, closely related to vegetation biomass.

Present approaches provide VOD as a soil moisture inversion residual at every time step and are therefore highly contaminated by residuals from model error. This work presents a novel technique for retrieving VOD using time-series of dual-polarized microwave observations. Taking advantage of the slow-time dynamics of VOD, a number of consecutive observations are used to estimate a single VOD. The soil dielectric constant of each observation is also retrieved simultaneously and later used as a consistency check. The method has been applied to two years of L-band passive observations from the NASA’s Aquarius sensor. Results show global VOD distribution follows general gradients of climate and canopy biomass conditions, with characteristic seasonal variability among the major land cover classes.

Satellite retrievals of microwave VOD provide independent but complementary information to other remote sensing vegetation metrics such as fluorescence and optical-infrared indices. The method presented here could be used in satellite missions such as SMOS and SMAP to decouple soil effects from vegetation, for the benefit of soil moisture retrievals. Also, it could be used to generate a new observational record of vegetation water content for a more comprehensive view of land surface phenology and terrestrial ecology.