Development of a Long-term Consistent Satellite Record of Soil Moisture

In recent decades, passive microwave remote sensing at low frequencies has become a primary means to measure and monitor the spatial and temporal variability of soil moisture on a global scale. Despite the development of individual specific radiometers (TRMM/TMI, Aqua/AMSR-E, Coriolis/WindSat, GCOM-W/AMSR2, and more recently GPM/GMI) with spectral specifications suitable to soil moisture estimation from space throughout the last two decades, there has not been a concerted effort to bring together the data from all these instruments to put into a long-term consistent record of soil moisture observations. Such a record would provide critical insights into the onset and evolution of long-term spatial and temporal patterns of soil moisture due to natural or anthropogenic changes in land surface conditions.

In this presentation, we will describe a scheme to produce a long-term consistent record of soil moisture from satellite observations as a Making Earth System Data Record for Use in Research Environments (MEaSUREs) investigation. This scheme begins with brightness temperature observations at X-band (10.7 GHz) and Ka-band (36.5 GHz) frequencies acquired from past and present satellites. Inter-calibration among satellites is accomplished by using GPM/GMI as the common calibration reference. The resulting inter-calibrated brightness temperatures are then subject to a physical inversion model and ancillary data for soil moisture retrieval. Because consistency is reinforced at the level of brightness temperature observations among satellites, the resulting record of soil moisture retrieval is expected to exhibit internal consistency with spatial and temporal characteristics governed by one common inversion model. Preliminary results, product delivery schedule, and deliverables out of the investigation will be discussed in this presentation.