| Abstract: |
Several active and passive microwave sensors are part of the current fleet of Earth remote sensing satellites and new ones will be added over the next years. Signals of these different sensors can have individual system-related characteristics due to variable designs (e.g. acquisition geometry or polarization), but observe the same soil and vegetation media on ground without significant time lag. Hence, active and passive microwave signals co-vary depending on the geophysical properties of the observed media, reflecting their respective dynamics over space and time. This means if the characteristic covariation patterns can be identified, radar and radiometer signals can be transferred into each other for joint data analyses and combined estimation of geophysical media properties.
Hence, covariation of remote sensing signals leads to meaningful (comprehensive and replicable) multi-sensor data integration and bases on the assumption that the utilized frequencies are sensitive to the same properties of the monitored media. Consequently, information about soil and vegetation properties, like their moisture status, can be assessed by combining active and passive microwave sensors, which unify the strength of the single-sensor approaches and jointly mitigate their individual weaknesses.
This methodology of covariation-based data integration is implemented by NASA’s Soil Moisture Active Passive (SMAP) Mission for surface soil moisture retrieval [1]. The baseline method rests on estimating active-passive covariation patterns from time series of radar and radiometer acquisitions at L-band (single-frequency (L/L) acquisition case) [2].
However, after the failure of SMAP’s L-Band radar in July 2015, its substitution with Sentinel-1’s C-band instruments for a combined active-passive retrieval of soil moisture demands an algorithm update for this dual-frequency (L/C) case. Due to the limited number of coincident SMAP and Sentinel-1 overpasses and the changing acquisition geometries of Sentinel-1, a time-series retrieval of active-passive covariation patterns was no longer feasible. Thus, efforts were needed to set up a novel algorithm having single-pass capabilities.
In this research contribution, active-passive microwave covariations of vegetated soils are modeled using a fully physics-based approach, which is applicable for the single- as well as the dual-frequency acquisition case. The methodology is based on Kirchhoff’s law of energy conservation and enables a detailed analysis of the dependency between active-passive microwave covariation and governing physical variables, like for instance surface roughness or vegetation water content [3]. In addition, dependencies between active-passive covariation and the acquisition scenario, like incidence angle and polarization, are analyzed and understood.
Moreover, the inversion of the physics-based forward model is presented, pointing towards a single-pass retrieval methodology for microwave covariation.
[1] Entekhabi, D., Njoku, E.G., O'Neill, P.E., Kellogg, K.H. Crow, W.T., Edelstein, W.N., Entin, J.K., Goodman, S.D., Jackson, T.J., Johnson, J., Kimball, J., Piepmeier, J.R., Koster, R.D., Martin, N., McDonald, K.C., Moghaddam, M., Moran, S., Reichle, R., Shi, J.-C., Spencer, M.W., Thurman, S.W., Tsang, L. & Van Zyl, J.J., (2010): “The Soil Moisture Active Passive (SMAP) Mission,” Proceedings of the IEEE, 98, 704-716.
[2] Entekhabi, D., Das, N.N., Njoku, E.G., Johnson, J., Shi, J., (2012): “Soil Moisture Active Passive: Algorithm Theoretical Basis Document L2 & L3 Radar/Radiometer Soil Moisture (Active/Passive) Data Products,” JPL Report, Jet Propulsion Laboratory, Pasadena, CA, USA.
[3] Jagdhuber, T., Konings, A.G., McColl, K.A., Alemohammad, S.H., Das, N.N., Montzka, C., Link, M., Akbar, R., Entekhabi, D, (2017): ”Physics-Based Modeling of Active and Passive Microwave Covariations Over Vegetated Surfaces,” IEEE Transactions on Geoscience and Remote Sensing, in review, 2017.
|
