satellite soil moisture observations for hydrological modelling [6–8, 20–22], with
some correlations explored between the satellite soil moisture datasets and the
hydrological models’ soil moisture state variables, their results have limited success
and could be improved further. One possible way is by analysing the fundamental
differences between the hydrological model-simulated soil moisture and the
satellite-measured soil moisture, so that the satellite observations could be enhanced.
The motivation of this study is to review the existing literature and explore the
potential issues in current satellite soil moisture application in hydrological modelling, which is topical and timely.
2 Soil Moisture Measuring Methods
First, it is necessary to give a brief introduction of the existing main soil moisture
measuring methods, so that basic knowledge about soil moisture could be better
understood. The following are based on two major categories: in situ and satellite
remote sensing.
2.1 In Situ Instruments
There are several techniques for in situ soil moisture measurements. The most
widely used ones are tensiometer, neutron probe, and time domain reflectometry
(TDR). Tensiometer is widely used in irrigation scheduling to help farmers to
identify the optimal time for irrigation [23]. It is also useful for plant studies.
However this approach is not suitable for sandy soil due to a limited range of bar
(0–0.8) and is not electronically stable for automatic operation [24]. For neutron
probe, although it is able to measure soil moisture at multi-depths fairly quickly and
automatically, it is not capable of giving reliable estimation at shallow depths
because some neutrons can escape from the soil surface into the air. Moreover,
since the device includes radioactive material, its operation requires extremely strict
training and inspection processes [25]. For TDR, the advantages of this method
are its high accuracy, fast response, free from radiation hazard and automatic soil
moisture estimation [26]. However the calibration of the sensor can be difficult and
expensive, and the instrument is easy to corrode [27].
Apart from the aforementioned point-based soil moisture estimation methods,
a novel technique capable of measuring area-averaged soil moisture has been
introduced. It is called COsmic-ray Soil Moisture Observing System (COSMOS).
The working mechanism of COSMOS is it measures the cosmic-ray neutrons above
the ground, whose intensity is primarily dependent on soil moisture [28]. One
COSMOS sensor can cover a horizontal effective area of about 600 m diameter
[29] and the measurement depth from about 12 cm (wet soil) to about 76 cm (dry soil)
[28, 30]. The COSMOS has been mainly installed in the USA and the UK as shown
Satellite Remote Sensing of Soil Moisture for Hydrological Applications. . .
261
some correlations explored between the satellite soil moisture datasets and the
hydrological models’ soil moisture state variables, their results have limited success
and could be improved further. One possible way is by analysing the fundamental
differences between the hydrological model-simulated soil moisture and the
satellite-measured soil moisture, so that the satellite observations could be enhanced.
The motivation of this study is to review the existing literature and explore the
potential issues in current satellite soil moisture application in hydrological modelling, which is topical and timely.
2 Soil Moisture Measuring Methods
First, it is necessary to give a brief introduction of the existing main soil moisture
measuring methods, so that basic knowledge about soil moisture could be better
understood. The following are based on two major categories: in situ and satellite
remote sensing.
2.1 In Situ Instruments
There are several techniques for in situ soil moisture measurements. The most
widely used ones are tensiometer, neutron probe, and time domain reflectometry
(TDR). Tensiometer is widely used in irrigation scheduling to help farmers to
identify the optimal time for irrigation [23]. It is also useful for plant studies.
However this approach is not suitable for sandy soil due to a limited range of bar
(0–0.8) and is not electronically stable for automatic operation [24]. For neutron
probe, although it is able to measure soil moisture at multi-depths fairly quickly and
automatically, it is not capable of giving reliable estimation at shallow depths
because some neutrons can escape from the soil surface into the air. Moreover,
since the device includes radioactive material, its operation requires extremely strict
training and inspection processes [25]. For TDR, the advantages of this method
are its high accuracy, fast response, free from radiation hazard and automatic soil
moisture estimation [26]. However the calibration of the sensor can be difficult and
expensive, and the instrument is easy to corrode [27].
Apart from the aforementioned point-based soil moisture estimation methods,
a novel technique capable of measuring area-averaged soil moisture has been
introduced. It is called COsmic-ray Soil Moisture Observing System (COSMOS).
The working mechanism of COSMOS is it measures the cosmic-ray neutrons above
the ground, whose intensity is primarily dependent on soil moisture [28]. One
COSMOS sensor can cover a horizontal effective area of about 600 m diameter
[29] and the measurement depth from about 12 cm (wet soil) to about 76 cm (dry soil)
[28, 30]. The COSMOS has been mainly installed in the USA and the UK as shown
Satellite Remote Sensing of Soil Moisture for Hydrological Applications. . .
261