Two soil moisture parameters are derived from active microwave
measurements (ERS-1/2 scatterometer): the
topsoil moisture content (surface
wetness) which is the relative measure of soil moisture
in the first 5 cm of the soil ranging between 0 and 1 representing
the degree of saturation; the Soil Water Index
(SWI), which is a relative measure of the soil moisture content in
the first meter of the soil layer, ranging between 0 and 100. If soil
hydrologic soil properties are known (wilting level, field capacity and
total water holding capacity) the SWI can be related to the volumetric
soil moisture content. The content of water within the uppermost meter of soil
is also derived from passive microwave measurements (Aqua/AMSR).
ONC requires soil
moisture information as input of SVAT scheme for carbon cycle modelling.
OFM needs the
SWI to calculate the Crop Performance Index and the Crop Yield.
Vienna University of Technology (IPF)
and
University of Bonn assess soil moisture
parameters from active and passive micro-wave sensors measurements,
respectively.
Monthly mean global SWI derived from ERS
scatterometer for the years 1992-2000.
IPF retrieves soil moisture using an advanced change detection approach
(Wagner et al., 1999b) which fully exploits the sensor design of the
ERS scatterometers with three antennas simultaneously observing the
Earth surface at different look directions and the availability of 10
years of high quality backscatter data. A reference backscatter value
representing backscatter from the vegetated land surface under dry soil
conditions is subtracted from the actual incidence-angle normalized
measurements to account for roughness and heterogeneous land cover. In
Wagner et al. (1999a), the method has been refined to account for the
effects of plan growth and decay by exploiting the multi-incidence
capabilities of the ERS scatterometer. As a result, time series of the
topsoil moisture content ms (< 5 cm) are obtained.
It is a relative quantity ranging between 0 (dry) and 1 (saturated).
In order to retrieve soil moisture in the root zone (up to about one
meter) a two-layer water balance model, which only considers the
exchange of soil water between the topmost remotely sensed layer
and the "reservoir" below, was used to establish a relationship
between the ms series and the profile soil moisture
content (Ceballos et al., 2005). The resulting quantity is called the
Soil Water Index (SWI) and ranges between 0 (wilting level) and 1 (field
capacity). The dependency of the reflected signal to the incidence angle
depends on the amount of vegetation on the surface. For correcting
vegetation effects, we use the fact that there exist an incidence
angle, which varies with the moisture conditions, were the effect
of vegetation is minimized. Soil moisture is retreived using a linear
relationship with the vegetation corrected signal, the highest
(representing saturated soils) and the lowest measurements (representing
dry soils) ever recorded. For desert regions where saturated conditions
are potentially never observed, this relationship is artificially force.
Soil moisture derived from AMSR data,
central 10-day period of july 2003.
Bonn University investigates the
soil moisture and its variability by analysing ground measurements from
the former Soviet Union. The data set comprises soil moisture measurements
of the upper 1 meter soil layer at 50 stations. Calculations are
restricted to the period 1979 to 1985, in which all additional ancillary
data sets are available. The variance of soil moisture shows a
pronounced dominance (about 85%) of the spatial variability between
the long-time means at each station. Aiming at a soil moisture algorithm
for decadal and continental applications, a two
step methodology is applied (Drusch, et al., 2001): first, the
temporal constant soil moisture is calculated using long-time
precipitation from GPCP (Global Precipitation Climatology Project),
the vegetation density from UMD-1km land cover map, and soil texture
and terrain slope from FAO. A linear regression is used to
retrieve the local climatological mean constant soil moisture within
the uppermost meter. By the second step, the temporal variability
is added. The 18 Ghz brightness temperature from Aqua/AMSR is
used to estimate the remaining temporal variance of soil moisture
at each grid point. Finally, the complete soil moisture field is
retrieved by combining the results of the two steps.
Both algorithms and products have been validated by comparison with
extensive database of multi-year ground measurements. As the soil
moisutre datasets derived from ERS scatterometer and from Aqua/AMSR do
not averlap, a direct comparison id not possible. Then, a validation
study is under progress at IPF to compare both soil moisture datasets
to the in-situ measurements of the RHEMEDUS network in Central Spain
where soil moisture parameters are collected since 1999.
References
Wagner, W., G. Lemoine, M. Borgeaud, and H. Rott, A study of
vegetation cover effects on ERS scatterometer and soil data, Remote
Sensing of Environment, 37(2), 938-948, 1999a.
Wagner, W., G. Lemoine, H. Rott, A method for estimating soil
moisture from ERS scatterometer and soil data, Remote Sensing of
Environment, Vol. 70, pp. 191-207, 1999b.
Drusch, M., E. F. Wood, and T. Jackson, Vegetative and
atmospheric corrections for the soil moisture retrieval from passive
microwave remote sensing data: results from the Southern Great Plains
Hydrology Experiment 1997. Journal of Hydrometeorology, 2, 181-192,
2001.
