H43H-1623
Intercomparison of AMSR2 and AMSR-E Soil Moisture Retrievals with MERRA-L data set over Australia
Thursday, 17 December 2015
Poster Hall (Moscone South)
Eunsang Cho1, Minha Choi2, Chun-Hsu Su3, Dongryeol Ryu3, Hyunglok Kim4 and Jennifer M Jacobs5, (1)Hanyang University, Seoul, South Korea, (2)Organization Not Listed, Washington, DC, United States, (3)University of Melbourne, Parkville, VIC, Australia, (4)Sungkyunkwan University, Water Resources and Remote Sensing Laboratory, Department of Water Resources, Graduate School of Water Resources, Suwon, Gyeonggi-do, South Korea, (5)Univ New Hampshire, Durham, NH, United States
Abstract:
Soil moisture is an important variable in the hydrological cycle on the land surface and plays an essential role in hydrological and meteorological processes. The Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E) sensor on board the Aqua satellite offered valuable soil moisture data set from June 2002 and October 2011 and has been used in a wide range of applications. However, the AMSR-E sensor stopped operation from 4 October 2011 due to a problem with its antenna. AMSR-E was replaced by the Advanced Microwave Scanning Radiometer 2 (AMSR2) on the Global Climate Change Observation Mission 1 – Water (GCOM-W1) satellite in May 2012. Assessment of AMSR2 soil moisture retrievals as compared to AMSR-E has not yet been extensively evaluated. This task is critical if AMSR2 soil moisture products are used as a continuous dataset continuing the legacy of AMSR-E. The purpose of this study is to inter-compare AMSR2 and AMSR-E microwave based soil moisture over Australia, mediated by using model-based soil moisture data set to determine statistically similar inter-comparison periods from time periods of the individual sensors. This work use NASA-VUA AMSR2 and AMSR-E based soil moisture products derived by the Land Parameter Retrieval Model (LPRM) and the modelled soil moisture from NASA’s MERRA-L (Modern Era Retrospective-analysis for Research and Applications-Land) re-analysis. The satellite soil moisture products are compared against the MERRA-L using traditional metrics, and the random errors in individual products are estimated using lagged instrumental variable regression analysis. Generally, the results demonstrate that the two satellite-based soil moisture retrievals have reasonable agreement with MERRA-L soil moisture data set. The error differences are notable, with the zonal error statistics are higher for AMSR2 in all climate zones, though the error maps of AMSR2 and AMSR-E are spatially similar over the Australia regions. This study leads to a better understanding of the newly developed AMSR2 soil moisture and thus helps AMSR2 to extend the valuable legacy of AMSR-E predecessor.
