Abstract:
This paper shows results obtained by using SMOS Level 2 retrieval algorithm, run at prototype stage, over forests. For each SMOS pixel, the algorithm estimates the soil m...Show MoreMetadata
Abstract:
This paper shows results obtained by using SMOS Level 2 retrieval algorithm, run at prototype stage, over forests. For each SMOS pixel, the algorithm estimates the soil moisture (SM) and the vegetation optical depth (τ). Average τ values retrieved in 4 days of July 2011 over forest pixels are shown and compared against forest height estimated by GLAS Lidar on board ICEsat satellite. Results of the comparison show a significantly increasing trend of τ with respect to forest height. For each 1-m interval of forest height estimated by Lidar, the standard deviation of optical depth is slightly higher than 0.1. The analysis is made again considering forest τ retrieved in 4 days of February, May, and November 2011, and it is observed that seasonal effects over optical depth are low. As an insight, it is shown that the increasing trend is still observed after subdividing world forests into Coniferous, Deciduous Broadleaf, and Evergreen Broadleaf. Comparisons with independent information about biomass are also shown at regional level for the U.S. The increasing trend is still observed, but with a reduced range of values. For SM, 14 nodes of the SCAN/SNOTEL network in the U.S. are considered. Over 2 years of data, retrieved values of SM are compared against ground measurements. Obtained values of correlation coefficient, rms error, and bias are reported.
Published in: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing ( Volume: 7, Issue: 9, September 2014)