I. Data Group Title
Soil Moisture
II. Revision Date
May 16, 1994
III. Authors
Stephen Ungar, Blanche Meeson, K. F. Huemmrich
IV. Data Group Overview
The soil moisture data sets cover a range of techniques for acquiring
and compiling information about soil moisture over the FIFE site. The
measurements here include ground based measurements (gravimetric, neutron
probe, and soil impedance) and those from airborne sensors (gamma ray and
Push-Broom Microwave Radiometer (PBMR)). Soil moisture data were collected
to describe conditions at the FIFE sites using gravimetric and neutron probe
methods. Gravimetric data were also collected for comparison with the PBMR,
soil impedance, and gamma ray data along transects.
The gravimetric and neutron probe data were combined with the gamma ray data
and interpreted to produce a map of the soil moisture over the FIFE study
area. A series of soil moisture contours were produced, these data were
digitized into a GIS and gridded. The contour locations are provided on CD
#1 and the gridded product is available on CD #5.
These data provide information about the moisture content of the soils at the
surface and within the root zone of the vegetation within the FIFE study
area. In addition, these data provide proof-of-concept for new techniques
for measuring soil moisture using remote sensing or other indirect
techniques.
V. Data Set Abstracts
Peck Airborne Gamma Ray Soil Moisture
[Investigator: Eugene Peck]
This data set contains measurements of soil moisture collected from a
gramma ray sensor flown on the NOAA aerocommander aircraft. Flightlines of
data were collected over 24 transects for which ground based gravimetric
measurements of soil moisture to 20 cm had been previously made. This data
set contains soil mositure values in percent for segments of a flightline.
In some cases, multiple segments were averaged to produce the soil moisture
value reported in these data. These transects covered most of the FIFE site
on 15 days in 1987 and 1989 (June 2nd and 3rd, 1987; July 6-10, 1987; August
2, 1989 and August 4-10, 1989).
Peck Gravimetric Soil Moisture
[Investigator: Eugene Peck]
This data set contains measurements of gravimetric soil moisture for
sites along the transects used as the reference for the airborne gramma ray
soil moisture data summarized above. Along with the soil moisture the site
slope and aspect, and a description of the site are included in this data
set. Measurements were made on the same days as the overflights carrying the
airborne gramma ray instrument to calibrate the instrument.
Soil Moisture Transect
[Investigator: E.T. Engman]
Gravimetric soil moisture measurements were made along transects that
were also measured by the impedance probe and overflown by the PBMR.
Volumetric soil moisture, bulk density and soil temperature measurements as
well as the wet and dry weights used to compute the soil moisture are
reported in this data set.
Soil Impedance Measurements of Soil Moisture
[Investigator: Stephen Ungar]
Daily measurements of the soil dielectric properties at 5 and 10 cm
depths were obtained at five locations throughout the FIFE test site during
the 1987 Intensive Field Campaigns (IFC). In addition, impedance probe
measurements were made throughout each of the IFCs along transescts
underlying airborne Pushbroom Microwave Radiometer (PBMR) flights.
Neutron Probe Soil Moisture Profile
The neutron probe data present a series of estimates of volumetric water
content in the soil profile which ranged from 20 cm to 2 m. Neutron probe
measurements were made surrounding the centers of selected FIFE sites in 1987
and in the WABs in 1989. The neutron probe provides an estimate of the
volumetric water content of the soil based on the thermalization of neutrons
colliding with atomic nuclei in the soil. Because of the sphere of influence
of the neutron probe instrument would extend into the air above the soil
surface, measurements in the top 20 cm of the soil were determined
gravimetrically and converted to volumetric values using measurements of bulk
density.
Gravimetric Soil Moisture
Gravimetric measurements of soil moisture were taken regularly at many
FIFE sites. Data collection daily throughout the IFCs and at one to two week
intervals between IFCs. Data were also collected approximately weekly during
1988 and daily during the 1989 IFC. Soil moistures were measured using the
gravimetric technique were taken from 0 to 5 and 5 to 10 cm. Dry and wet
weight measurements as well as the soil moisture amounts derived from these
weights are provided in this data set.
Contours of Soil Moisture for the FIFE Research Area
[Investigator: Eugene Peck]
This data set is a set of contours for soil moisture for the entire FIFE
site for 15 days in 1987 and 1989. There are 7 maps for 1987 (two in early
June and five in early July) and 8 in early August 1989. The data set
contains the soil moisture contour value in percent and the locations for
that contour. These contours were compiled from data collected on the same
day using the airborne gamma radiation measurements, and the gravimetric and
neutron probe measurements.
Soil Moisture Grid Measurements
[Investigator: Peter M. Groffman]
Grid sampling of soil moisture (gravimetric) was done during FIFE 87 and
FIFE 89 to characterize the effects of topography on soil moisture
variability. Three sampling grids were located on the 1D watershed in flat
(F), slope (S) and valley (V) positions Grids were 66 [m^2], and were
divided into 49 sampling points, 11 meters apart. This data set is not in
FIFE standard format and is reported in the "Grab_Bag" directory of the CD-ROM.
FIFE Level-3 Example Gridded Soil Moisture Data
[Investigator: Allen Hope]
Soil moisture values were estimated for a grid of points located in the
FIFE research area for 7 days in 1987 and 8 days in 1989. These grid point
estimates are based on composite soil moisture maps prepared by combining
data from several sources, including soil maps, point data, remote
sensing, and model output products. The grid node values in this product
represent the average soil moisture of a 0.5 km x 0.5 km area centered at
the node location.
VI. Special Properties of These Data
Gravimetric techniques yield the soil moisture by weight of water per
unit weight of dry soil. In order to convert to volumetric soil moisture,
which can be related to the total water contained in a surface layer of
specified thickness, it is necessary to know the bulk density of the soil.
The airborne gamma ray instrument measures the gamma radiation emitted from
sources distributed in the top 10 to 30 cm of the soil. The actual depth
observed varies due to the attenuation of the soils to gamma rays. The
attenuation increases with increasing volumetric soil moisture. So to
calculate soil moisture the gamma ray data must be calibrated once using a
set of ground soil moisture measurements from along the flight line. There
is an atmospheric component to the gamma ray attenuation so it is important
to fly at a constant altitude.
The impedance probe measures the dielectric constant at 10 MHz within a fixed
cylindrical volume of soil. Since the dielectric constant of water is almost
two orders of magnitude greater that that of most dry soils, the dielectric
constant of the sample is extremely sensitive to the volumetric water
content.
The neutron probe provides an estimate of the volumetric water content of the
soil because of the hydrogen present in water significantly affects the
thermalization of neutrons colliding with atomic nuclei in the soil. The
thermal neutron density is calibrated against the water concentration on a
volume basis.
VII. Relations Among These Data
The FIFE staff collected gravimetric and neutron probe data at selected
FIFE sites on a coordinated basis. Soil impedance probe data is available
for five of the 1987 staff measurement sites and three "super" sites in 1989.
The Peck gravimetric data were taken along the gamma ray flight lines to
calibrate that data. The gravimetric data provide soil moistures for 0 to 20
cm. The soil moisture transect gravimetric data were used to support the
studies using the impedance probe and the Push Broom Microwave Radiometer
(PBMR; see the Satellite and Aircraft Image Data summary document on this CD-
ROM). The soil moisture transect data set was collected during PBMR
overflights along the same transects that were flown. In addition, gamma ray
flight data was acquired over the soil moisture transects, on selected days,
along wth PBMR, impedance probe and gravimetric data.
Soil impedance can be converted to soil moisture through two approaches.
Nominal volumetric estimates can be made from the measured dielectric
constant since the dielectric constant of pure water is much greater than
that of dry soil. Probe values can be calibrated by correlation with
overlapping gravimetric measurements (Ungar et al. 1992).
Other soil properties data were also collected in FIFE. These include
saturated soil hydraulic conductivity, soil thermal conductivity, bulk
density, and soil moisture release data. These data sets are described in
the Soil Properties summary document on this CD-ROM.
VIII. Confidence and Error
Both gravimetric and soil impedance probe data are point measurement data
subject to similar errors when used to typify conditions at specified
locations. In both cases the detailed determination of value is extremely
accurate and does not limit the accuracy of estimation. For the gravimetric
determination, the ability to accurately determine wet and dry weights as
well as to reproducibly dry the sample is very high and results in an
uncertainty of a fraction of a percent of the determined value. For the
probe, the electrical measurements used in determining the sample dielectric
constant are accurate to within 0.1 percent.
For both gravimetric and dielectric determinations the major sources of
estimation error are sampling techniques and small scale geographic
variations. Gravimetric coring samples may contain roots, rocks and voids.
Coring samples from spots as little as 10 cm apart can produce samples which
typically differ by up to 30 percent of each other's value. This is due to
both differences in the extraction of the samples and differences in the
drainage characteristics (e.g. one of the coring sights may have an
underlying rock causing pooling or roots in the proximity drawing moisture
away). The dielectric probe is, of course, subject to the same locational
sampling error. In addition, variations in the manner in which the probe is
inserted can lead to measurement differences amounting to several percent.
To minimize the uncertainty of gravimetric errors, it is necessary to take
several measurements (typically 3 to 5) to get a truly representative value
for a local estimate. This can be highly destructive for a repetitive
measurement scenario. For the FIFE, five to 10 gravimetric measurements per
site were taken each measurement day to typify the site average.
Neutron probe measurements are generally considered unreliable near the
surface (first 10 to 20 cm). Therefore neutron probe data was not collected
above 20 cm. To complete the soil moisture profile, the data set contains
gravimetric data converted to volumentric water contents for depths less than
20 cm. The different methods may cause inconsistencies in profiles. A
possible source of sampling error with the neutron probe technique results
from the accumulation of a moisture layer along the surface the neutron
activation tube. The effect of this layer on the soil moisture estimate is
difficult to quantify, but could amount to an error of several percent
absolute volumetric soil moisture.
A major problem with airborne soil moisture measurements (PBMR, Gamma Ray
Instrument, Scatterometer) is that the depth of the measurement is
indeterminate. The PBMR and Gamma Ray Instrument both sense radiation that
is attenuation limited and measure signals that involve anywhere from the top
5 to 25 cm of soil depending on the amount of moisture (and in the case of
the Gamma, distribution of sources) present.
The soil moisture contour data set was based on the data sets described above
and inherits all of the uncertainties contained therein. In addition, many
poritions of the FIFE Research Area are not well sampled by these data sets
and the contour values determined are the result of interpolation and
extrapolation
IX. Results
Soil moisture is a critical factor in understanding the surface radiation
balance. In FIFE, when the gravimetric soil moisture levels dropped below
approximately 20% rapid decreases in surface evaporation were observed as
plant stomata responded to conserve plant moisture. At soil moisture levels
greater than field capacity (approximately 40% gravimetric soil moisture),
soil evaporation began to play an important role in total surface
evapotranspiration (Sellers and Hall, 1992).
FIFE represented the first large scale field experiment that allowed the
simultaneous measurement of soil moisture variations using several different
approaches. Soil moisture measurements using the gamma ray technique showed
a very good agreement between the estimated and measured soil moistures with
the vegetation cover having little effect. The gamma ray technique has some
disadvantages in its coarse spatial resolution (at an altitude of 150 m the
gamma ray data have a pixel size of 300 m by 300 m) and the requirement for
extensive ground measurements for calibration. Brightness temperatures from
the PBMR showed a relationship with volumetric soil moisture except for
unburned areas in which a thick layer of thatch had developed. The thatch
layer is a good absorber of microwave emissions even at low frequencies.
The results of the impedance probe study indicate its usefullness in
monitoring soil moisture. The probe can be installed permanently at a
location and give soil moisture data without site disturbance. Impedance
probe measurements were made throughout each of the IFCs along transescts
underlying airborne Pushbroom Microwave Radiometer (PBMR) flights and
compared with the results of gravimetric sampling done in support of these
flights. Examination of the data revealed that the impedance probe is a more
consistence source of time series information than traditional measurements,
and is potentially more closely linked to the physical parameters which are
both remotely sensible and required for surface energy/mass exchange
determination.
X. Partial Bibliography
Peck, E. L., T. R. Carroll, and D. M. Lipinski, 1992, Airborne soil moisture
measurements for FIFE, 1992, J. Geophys. Res. 97(D17):18961-18968.
Sellers, P. J. and F. G. Hall, 1992, FIFE in 1992: results, scientific gains,
and future research directions, J. Geophys. Res. 97(D17):19091-19109.
Ungar, S. G., R. Layman, J. E. Campbell, J. Walsh, and H. J. McKim, 1992,
Determination of soil moisture distribution from impedance and gravimetric
measurements, J. Geophys. Res. 97(D17):18969-18978.
Wang, J. R., 1992, An overview of the measurements of soil moisture and
modeling of moisture flux in FIFE, J. Geophys. Res. 97(D17):18955-18960.
|
