Through reviewing various satellite techniques for soil moisture estimation,
hydrological evaluation of satellite soil moisture and building satellite soil moisture
error model, a major problem in satellite soil moisture utilisation for hydrological
modelling is concluded. It is that current satellite soil moisture products are mainly
calibrated by in situ soil moisture observations. As a result, they are not directly
relevant to catchment hydrological modelling. Therefore, a soil moisture product
that can be directly linked with hydrological models is desired, and more studies
about developing new hydrological soil moisture products as described in Sect. 6 are
needed. Certainly, there are still many specific challenges remained, for example,
currently the soil moisture retrievals are only available at a coarse spatial resolution
which could lead to spatial mismatch problems (e.g. over- or under-representation
of soil moisture condition of a catchment); therefore accurate retrieval up to a finer
resolution may be necessary particularly for those small-sized catchment studies.
This could be achieved by downscaling methods using machine learning techniques.
Another issue is about the depth mismatch between the hydrological model soil layer
and the satellite’s penetration. Currently for most operational hydrological models, it
is difficult to have fixed soil depths. On the other hand, satellite-retrieved data also
have unfixed soil depth, which depends on many factors such as vegetation and soil
roughness. The important thing is to create a linkage between a hydrological model’s
soil moisture state variable and the satellite data (e.g. build mathematical relationships). If a good linkage is built, the satellite soil moisture product can then be used
effectively in hydrological model’s soil moisture state initiating and updating during
real-time flood forecasting. Only breakthrough in those areas will lay a solid
foundation for future data assimilation of soil moisture observations in the realtime flood forecasting. Furthermore, it is hoped this study will attract attention from
the hydrological community on those problems and encourage more research to
solve them in a wide range of geographical and climatic conditions.
References
1. Prigent C, Aires F, Rossow WB, Robock A (2005) Sensitivity of satellite microwave and
infrared observations to soil moisture at a global scale: relationship of satellite observations to
in situ soil moisture measurements. J Geophys Res D Atmos 110(D7)
2. Ochsner TE, Cosh MH, Cuenca RH, Dorigo WA, Draper CS, Hagimoto Y, Kerr YH,
Njoku EG, Small EE, Zreda M (2013) State of the art in large-scale soil moisture monitoring.
Soil Sci Soc Am J 77(6):1888–1919
3. Aubert D, Loumagne C, Oudin L (2003) Sequential assimilation of soil moisture and
streamflow data in a conceptual rainfall–runoff model. J Hydrol 280(1):145–161
4. Christian RJ (1997) Validation and intercomparison of different updating procedures for realtime forecasting. Nord Hydrol 28(2):65–84
5. Ottlé C, Vidal-Madjar D (1994) Assimilation of soil moisture inferred from infrared
remote sensing in a hydrological model over the HAPEX-MOBILHY region. J Hydrol
158(3):241–264
Satellite Remote Sensing of Soil Moisture for Hydrological Applications. . .
275
hydrological evaluation of satellite soil moisture and building satellite soil moisture
error model, a major problem in satellite soil moisture utilisation for hydrological
modelling is concluded. It is that current satellite soil moisture products are mainly
calibrated by in situ soil moisture observations. As a result, they are not directly
relevant to catchment hydrological modelling. Therefore, a soil moisture product
that can be directly linked with hydrological models is desired, and more studies
about developing new hydrological soil moisture products as described in Sect. 6 are
needed. Certainly, there are still many specific challenges remained, for example,
currently the soil moisture retrievals are only available at a coarse spatial resolution
which could lead to spatial mismatch problems (e.g. over- or under-representation
of soil moisture condition of a catchment); therefore accurate retrieval up to a finer
resolution may be necessary particularly for those small-sized catchment studies.
This could be achieved by downscaling methods using machine learning techniques.
Another issue is about the depth mismatch between the hydrological model soil layer
and the satellite’s penetration. Currently for most operational hydrological models, it
is difficult to have fixed soil depths. On the other hand, satellite-retrieved data also
have unfixed soil depth, which depends on many factors such as vegetation and soil
roughness. The important thing is to create a linkage between a hydrological model’s
soil moisture state variable and the satellite data (e.g. build mathematical relationships). If a good linkage is built, the satellite soil moisture product can then be used
effectively in hydrological model’s soil moisture state initiating and updating during
real-time flood forecasting. Only breakthrough in those areas will lay a solid
foundation for future data assimilation of soil moisture observations in the realtime flood forecasting. Furthermore, it is hoped this study will attract attention from
the hydrological community on those problems and encourage more research to
solve them in a wide range of geographical and climatic conditions.
References
1. Prigent C, Aires F, Rossow WB, Robock A (2005) Sensitivity of satellite microwave and
infrared observations to soil moisture at a global scale: relationship of satellite observations to
in situ soil moisture measurements. J Geophys Res D Atmos 110(D7)
2. Ochsner TE, Cosh MH, Cuenca RH, Dorigo WA, Draper CS, Hagimoto Y, Kerr YH,
Njoku EG, Small EE, Zreda M (2013) State of the art in large-scale soil moisture monitoring.
Soil Sci Soc Am J 77(6):1888–1919
3. Aubert D, Loumagne C, Oudin L (2003) Sequential assimilation of soil moisture and
streamflow data in a conceptual rainfall–runoff model. J Hydrol 280(1):145–161
4. Christian RJ (1997) Validation and intercomparison of different updating procedures for realtime forecasting. Nord Hydrol 28(2):65–84
5. Ottlé C, Vidal-Madjar D (1994) Assimilation of soil moisture inferred from infrared
remote sensing in a hydrological model over the HAPEX-MOBILHY region. J Hydrol
158(3):241–264
Satellite Remote Sensing of Soil Moisture for Hydrological Applications. . .
275