A: GEM-300 B: Aqua Probe C: TDR

rawfrogpondUrban and Civil

Nov 16, 2013 (3 years and 10 months ago)

92 views


5

3. Methods


3.1 The EM technique and its applicability to measure soil moisture

The EM instruments measure the bulk electric conductivity. The latter is a function of the soil structure,
texture and mineralogy as well as its water content and the salt cont
ent of the water (Cook 1998). All
these factors are important, because the electrical currents are transferred through soil on different
paths. Electrons can move through the soil moisture, along the surfaces of particles that are in contact
with one anoth
er or through alternating layers of particles and solution (McNeill 1980a).


The general premise behind the use of GEM for soil moisture measurements is that given all other
parameters remain unchanged, the variability of the GEM measurement of the electri
c conduction with
time reflects the change of soil moisture with time. The temporal changes of the water content should
therefore be an easy a parameter easy to capture.

Combined with other soil moisture measurements in the same area or an accurate estimat
ion of the
amount of dissolved salt in the soil water and the clay content together with the conductivity of the
solid soil even an absolute calibration of EM readings for the soil moisture content of a certain area
should be possible
.


3.1.1 The relations
hip between bulk electric conductivity and soil moisture content

The measured conductivity is called the apparent or bulk conductivity and it is a function of the
combination of the different conducting soil components mentioned above.

Rhoades et al (1976)

show that the bulk electric conductivity is a quadratic function of the volumetric
water content:

EC
a
EC
b
EC
EC
a
w
v
v
w
c



*
*
*


2






(1)


where

v
is the volumetric water content, EC
w

is the electrical conductivity of the soil water, EC
c

is the
surface ele
ctric conductivity it is dependent on the type and amount of clay in the soil and
a

and
b

are
regression parameters related to soil texture.





6

Nadler and Frenkel (1980) and Shainberg (1980) suggest for conductivity above 400 mS/m equation
(1) can be appro
ximated with a linear equation:


EC
EC
T
EC
a
w
v
s


*
*









(2)


where EC
w
is the electrical conductivity of the soil solution,

v

is the volumetric water content, T is the
tortuosity and EC
s
is the solid phase contribution to the apparent electri
cal conductivity.

One goal of this experiment was to determine whether a linear or a quadratic relationship between
measured soil moisture and EM readings existed at the same measurement locations.


3.1.2 Basics of frequency
-
domain EM methods

The EM cre
ates a primary electromagnetic field which induces eddy currents in the subsurface
material. These in turn induce a secondary electromagnetic field, which modifies the total field. The
difference between these two fields is a function of the subsurface con
ductivity (Cook 1998).

The EM records the magnetic intensity of the secondary field that is given by the equation (Mares
1984):

H
t
H
t
(
)
cos(
)


0










(3)

where H
0

is the primary magnetic field,


is the angular frequency and the angle


deter
mines the
phase delay. The phase delay is caused by the short amount of time the particles in the ground need to
adjust to the imposed field (White 1994). The quantities which display a zero phase delay are referred
to as in
-
phase quantities and the quanti
ties which are displaced in phase with respect to the source
current by 90


are referred to as in quadrature. The apparent conductivity is therefore related to the
measured quadrature phase values (Won 1996). The relationship is linear.


Cond = 360*PPM(Q)/

Freq








(4)


Cond is the conductivity in units of mS/M

PPM(Q) is the quadrature phase reading

Freq is the frequency in Hz


7


This relationship in equation 4 is valid only for surveys over poorly conductive ground with an
induction number smaller than on
e. The induction number is defined as intercoil spacing (1.67m for the
GEM
-
300) divided by the skin depth. The skin depth can be derived from a special nomogram (Won
1996). For all the measurements that we have been conducting the induction number was <1.


3.1.2.1 The depth response

The penetration depth of the GEM ranges from zero up to several meters. The penetration depth of the
electromagnetic field is dependent on the frequency. Lower frequencies reach greater depths than
higher frequencies, due to the

stronger attenuation of the high frequency waves
(Won, 1996).

The
option of measuring different frequencies in the same cycle, potentially allows to create a 3D
-

image
of the soil water distribution. To date the depth response of the one frequency EM had

to be calculated
theoretically. The conductivity measured with the EM is the integrated conductivity over a specific
depth range. The influence of the conductivity of each depth each layer is given by a depth response
function

(z) (McNeill, 1980).

EC
z
EC
z
dz
a




(
)
(
)
0








(5)


where

(z) is the depth response function and EC(z) is the electric conductivity of the layer z. At low
soil conductivity, for example for non
-
saline soils, the response function is independent of the earth
conductivity.

McNeill (1980) calculated a

(z) function for the EM31 which operates at a frequency of
14 k Hz (Fig. 4) . This function should approximately be valid for the frequency f4 = 14010 Hz used in
the present study.


Fig. 3: horizontal

h
and vertical

v

dipole

response function ( McNeill, 1980)


8

If the

(z) function for a certain frequency is known, the conductivity distribution with z can be
calculated numerically. We assume that the total conductivity is a linear combination of the
conductivities for distin
ct soil layers. It is then possible to determine the conductivity of a distinct
layer by minimizing the magnitude of the

(z) below and above this certain the depth interval. This is
done by introducing a weighting factor
a




(
)
(
)
z
a
z
i
i
n
i



1








(6)



(z) is the weighted depth response function,
a
i


is the weighting factor for the
i
th layer and

i
(z) is the
response function for the
i
th layer.

With a good selection of
a

one can
damp

(z) below and above a certain
depth enough to only
receive a strong
signal from this distinct z layer (Cook,
1992).




Fig.4: Cummulative depth response functions
for different frequencies (Cook, 1998)




3.1.2.2 The GEM
-
300

For all experiments we used the new GSSI GEM
-
300. It consists of transmitter coil

with an alternating
current running through it, inducing an electromagnetic field at different frequencies. Seven different
frequencies can be registered within the same measuring cycle. The GEM can transmit frequencies
within the range from 325 up to 199
75 Hz. There is a receiver unit that records the in
-
phase and
quadrature components of the returning signal.




9

The data for both phases is converted into a PPM unit as follows: (Won 1996 )


PPM
H
H
s
p

10
6
*









(7)

Where H
s
is the second
ary magnetic field at the receiver coil and H
p

is the primary field at the receiver
coil.


A control processor unit runs the system software manages the power and stores temporarily the
measured data. The measured values can easily be transferred onto a co
mputer and then into a plotting
program such as ‘Magmap’ or ‘Surfer’ to display the data as 2D contour maps for further analysis.

The GEM
-
300 offers different survey modes. One can either perform a continuous survey or station
wise measurements. Both mode
s are applicable to all user
-
defined grids. The continuous mode is a very
fast way to do a survey with a flexible grid that needs only the grid end points to be laid out. The
‘station wise’ survey offers the option of very accurate positioning of the measu
rement points within a
user
-
defined grid.





A: GEM
-
300 B: Aqua Probe C: TDR

Fig. 5: The different measuring instruments


3.2 How the true moisture content was measured

Subsurface moisture was determined using vertical TDR and a depth integrated capacitance sensor to

The TDR consists of 20cm long stainless rods inserted vertically in the ground. An electric impulse
was then
sent along these two rods and compared to the returned signal.
The type of material
surrounding the two rods influences the returning signal. If the dielectric constant of the material is
high, the signal propagates slower. The dielectric constant of water

is much higher than most other
materials; therefore a signal within a wet or moist medium propagates slower than if the same medium

10

were dry. Moisture content can be determined by measuring the propagation time over a fixed length
probe

(Slik, 1993).

From

the time the dielectric constant can be derived as given in equation (8)












ct
L
p
2








(8)

where L
p

is length of the probe in the soil and t is the one way travel time of the wave along the rod,

To
pp (1980) linked the dielectric constant to the volumetric water content:















5
3
10
2
92
10
5
5
10
4
3
10
2
2
4
2
6
3
.
*
.
*
.
*
.
*



(9)

where


is the dielectric constant of the porous medium.

In our study TDR yielded a point value of the volumetric moisture content integrated over

the upper
20cm of the soil.

The Aqua Probe Moisture Probe used in this study is classified as a "Capacitance" sensor.


The
Moisture Probe transmits a very low powered "Radio Frequency" through the soil at user
-
defined depth
(0
-
80cm) to measure moisture.


There are two copper bands on the end of the sensor that function as
radio antennas.


One antenna transmits a low powered "RF" signal that is received by the other
antenna.


The proprietary microprocessor can determine the moisture content by the change in

frequency of the signal it receives
(link 3).

The more moisture in the soil, the more the frequency of the
signal is changed.


The AP has a repeatable accuracy of


+/
-


2% (link 3). The AP has a resolution of
1% of the relative water content. The AP prob
e yields relative soil moisture of the upper 50
-
80 cm of
the soil with a user defined resolution. A calibration with soil samples is needed to convert the relative
soil water content into the volumetric moisture content.


11