Transcript Methods of Media Characterization

Methods of Media
Characterization
A challenging area of
rapid advancement
1
http://www.ianr.unl.edu/pubs/irrigation/graphics/g690-06.jpg
Topics
Measurement of pressure potential
The tensiometer
The psychrometer
Measurement of Water Content
TDR (dielectric)
Neutron probe (thermalization)
Gamma probe (radiation attenuation)
Gypsum block (energy of heating)
http://www.civag.unimelb.edu.au/~jwalker/
data/nerrigundah/connector.jpg
http://www.unidata-starlog.de/
produkte/5513c.jpg
Measurement of Permeability
Tension infiltrometer
Well permeameter
http://www.wateright.org/
site2/images/neutron.jpg
2
Physical Indicators of Moisture
All methods measure some physical quantity
What can be measured?
weight of soil
pressure of water in soil
humidity of air in soil
scattering of radiation that enters soil
dielectric of soil
resistance to electricity of soil
texture of soil
temperature/heat capacity of soil
Each method takes advantage of one indicator
3
Methods: Direct versus indirect
Direct methods measures the amount of
water that is in a soil
Indirect methods estimates water content by
a calibrated relationship with some other
measurable quantity (e.g. pressure)
We will see that the vast majority of tools
available are “indirect”
The key to assessing indirect methods is the
quality/stability/consistency of calibration
4
Methods: direct
Gravemetric
Dig some soil; Weigh it wet; Dry it;
Weigh it dry
Volumetric
www.geog.plym.ac.uk/
labskills/bdpg.htm
Take a soil core (“undisturbed”); Weigh wet, dry
Pro’s
- Accurate (+/- 1%)
- Cheap
equipment - free
per sample - free
Con’s
- Can’t repeat in spot
- Slow - 2 days
- Time consuming
5
Methods: Indirect via pressure
http://www.ci.eagan.mn.us/Forestry/
6_1_01_tensiometer.jpg
Tensiometers
Psychrometers
2
Indirect : Surrogate media
Gypsum blocks (includes

WaterMark etc.)
6
Communicating with soil:
Porous solids
The tensiometer
employs a rigid
porous cup to allow
measurement of the
pressure in the soil
water.
Water can move freely
across the cup, so
pressure inside is
that of soil
7
Pressure measurement:
The tensiometer
Can be made in many
shapes, sizes.
Require maintenance to
keep device full of water
Useful to -0.8 bar
Employed since 1940’s
Need replicates to be
reliable (>4)
Reservoir
Gauge
Body
Removable
Cup
8
Pressure measurement: The tensiometer
Can be made
in many
shapes, sizes.
9
Pressure measurement:
The tensiometer
Thumbnail: Watch out for:
Swelling soils
tensiometer will loose contact, and not function
Inept users!
 Poor for sites with low skill operators of units
 Easy to get “garbage” data if not careful
Fine texture soils (won’t measure <-0.8bar)
Most useful in situations where you need
to know pressure (engineered waste etc.)10
Pressure potential: The psychrometer
A device which allows
determination of the relative
humidity of the subsurface through
measurement of the temperature of
the dew point
Relative humidity
Gas Temperature
constant
Pressure
 Pg 
Pd = R T ln 
 P 
nevada.usgs.gov/adrs/ pg_hydro.html
Porous Ceramic Cup
_
Thermocouple
11
Pressure potential: The psychrometer
Thumbnail: most likely not your 1st choice...
Great for sites where the typical conditions
are very dry. In fact, drier than most plants
prefer.
Low accuracy in wet range (0 to -1 bar)
Need soil characteristic curves to translate
pressures to moisture contents - problem in
variable soils
Great for many arid zone
research projects
12
http://www.decagon.com/wp4/
http://www.dynamax.c
om/gypsum.jpg
Indirect pressure: Gypsum
block, Watermark et al.
 Using a media of known moisture
content/pressure relationship
 Energy of heating a strong function of 
 Resistance embedded plates also f().
 Measure energy of heating, or resistance;
infer pressure
http://www.unidata-starlog.de/
produkte/5513c.jpg
Thermocouple
Heater
Problems:
 The properties of the media change with
time (e.g., gypsum dissolves; clay
deposition)!
 Making reproducible media very difficult
(need calibration per unit)
 Hysteresis makes the measurement
inaccurate (more on this later)
W
13
Example: Watermark
$260 for meter
$27 for probes
http://www.irrometer.com/images/watsensor.JPG
Indirect Pressure:
Gypsum block, Watermark et al.
Idea of indirect pressure measurements:
Measure water content of surrogate media, infer pressure,
then infer water content in soil
Surrogate
Media
Water content
We measure water content
in the surrogate media
Soil
Water content
We want a value for water
15
content in our soil
Indirect Pressure:
Gypsum block, Watermark et al.
Thumbnail:
Generally a low cost option
Calibration typically problematic in time
and between units
Poor in swelling soils
Poor in highly variable soils
Sometimes adequate for yes/no
decisions
We have had very poor luck with these
16
in Willamette valley (no correlation!)
Indirect electrical:
the nature of soil dielectric
• Soils generally have a dielectric of
about 2 to 4 at high frequency.
• Water has a dielectric of about 80.
• If we can figure a way to measure the
soil dielectric, it shows water content.
WATCH OUT: the soil dielectric is a
function of the frequency of the
measurement! For it to be low, need to
use high frequency method (>200 mHz)
17
Indirect electrical: Capacitance (dielectric,
low frequency)
Stick an unprotected
capacitor into the soil
$70
and measure the
capacitance.
Higher if there is lots of
dielectric (i.e., water)
Need to Calibrate capacitance
vs volumetric water content
per soil
$500
PROBLEM:
soils have very different
dielectrics at low frequency
18
19
High Frequency Capacitance (Dielectric)
80 mHz
$250 meter
$250 sensor
$20 access tube
Calibration fairly
stable
20
Indirect electrical: TDR (dielectric)
Observe the time of
travel of a signal
down a pair of wires
in the soil.
Signal slower if there
is lots of dielectric
(i.e., water)
Calibrate time of
travel vs volumetric
water content
Since high frequency,
can use “universal”
calibration
21
Indirect electrical: TDR (dielectric)
Lots of excitement surrounding TDR now. Why?
Non-nuclear
universal calibration
measures volumetric water content directly
wide variety of configurations possible
Long probes (up to 10 feet on market)
Short probes (less than an inch)
Automated with many measuring points
Electronics coming down in price (soon <$500)
Potentially accurate (+/- 2% or better)
22
Indirect radiation:
interactions between soil & radiation
When passing through, radiation can
either:
be adsorbed by the stuff
change color (loose energy)
pass through unobstructed
http://www.pnl.gov/flowcells/images/satunsat2.jpg
Which of these options occurs is a
function of the energy of the radiation
Each of these features is used in soil
water measurement
23
Indirect radiation:
Neutron probe (thermalization)
Send out high energy neutrons
When they hit things that have same mass as a neutron
(hydrogen best), they are slowed.
Return of slow neutrons calibrated to
water content (lots of hydrogen)
Single hole method
(access tube)
Quite accurate (simply
wait for lots of counts)
Lots of soil constituents
can effect calibration
http://www.wateright.org/
site2/images/neutron.jpg
High Energy Neutrons Thermalized Neutrons
24
Indirect radiation:
Neutron probe (thermalization)
Cons
Pro’s
Potentially Accurate Needs soil specific
calibration (lots of
work)
Widely available
Working with radiation
Inexpensive per
Expensive to buy
location
Expensive to dispose
Flexible (e.g., can go
Slow to use
very deep)
can’t be automated
25
Indirect radiation: Gamma probe
Sample
Radiation attenuation
Source & detector separated
by soil.
Source
Water content determines
adsorption of beam energy.
Must calibrate for each soil.
Same used in neutron and
x-ray attenuation.
Can use various frequencies
to determine fluid content of
various fluids (e.g., Oils)
Not used in commercial agriculture
Detector
http://www.pnl.gov/flowcells/images/satunsat2.jpg
26
Gamma Attenuation
Attenuation follows Beer’s law:
each frequency attenuated at
different rate; each material having
a different attenuation rate.
I= incident radiation
I= transmitted radiation
 xi=thickness of medium i
 ai=attenuation coefficient
for material i at
frequency 
I /I
o =
exp(
Sample
Source
i x i ) = C exp( wx w)
i
Detector
[2.166]
27
Indirect via feel:
getting to know your soil
Soil water status obtained
checking the feel of the soil
Does It make a ribbon?
Does it stick to your hand?
Does it crumble?
Although crude, the information
immediate; gets farmer in field
thinking about water and her soil
Possibly the most effective water monitoring
strategy
http://www.ianr.unl.edu/pubs/
irrigation/graphics/g690-06.jpg
28
Directions in the future
Much lower cost TDR
Much more flexible systems
radio telemetry for cheap
auto-logging systems
computer based tracking
http://www.historyoftheuniverse.com/
images/future.gif
Much less water to work with
Much more call for precise and frequent water
monitoring
29
Permeability: Double ring infiltrometer
Establishes 1-d flow
by having concentric
sources of water
measure rate of
infiltration in central
ring
Easy, but requires
lots of water, and
very susceptible to
cracks, worm holes,
etc.
Interogates large
area
Typical Double Ring Infilrometer Set-up
Constant head must be
maintained, to be equal in inner and
outer rings
Install deep
enough to avoid
leakage
Infiltration under the
inner ring is approximately
one dimensional
30
Available in a Wide Range of Sizes!
http://www.geo.vu.nl/~geomil/pics/reading-ring-infiltrometer-small.jpg
Photo: Paul Measles
http://www.gw-env-group.com/Photos/Geotechnical/Double_Ring_Test.jpg
31
http://www.turf-tec.com/in10-w.jpg
Interpreting Infiltration Experiments
Horton Equation:
Rate of infiltration, i, is given by
i = if + (io - if) exp(-t)
where if is the infiltration rate after long time, io
is the initial infiltration rate and  is and
empirical soil parameter. Integrating this with
time yields the cumulative infiltration
io - i
I =i ft +

f
( 1 - exp(
t ) )
32
The Brutsaert Model
The Brutsaert Model
i = K sat + 0.5 S t
-1/2

1 +

Ksat t
 
S

1/2
 
 
 
-2
S = sorptivity
0<<1 pore size distribution parameter. wide pore
size distributions  = ;1 other soils  = 2/3
The Brutsaert cumulative infiltration is
I = K sa t

S2 
t+
1  K sa t 


Ksa t t
 1+  
S


1/2
 
 
 
-1 




from which you can determine Ksat and S.
33
Interpreting Infiltration Experiments, cont.
The two term Philip model suggests fitting the
rate of infiltration to
i = 0.5 S t-1/2 + A
and the cumulative infiltration as
I = S t1/2 + At
34
Interpreting Infiltration Experiments, cont.
The Green and Ampt Model (constant head)
n 
t=
L - (h
Ksat 
h f +d+L  
f +d) ln  h +d  
 f
 
L = depth of wetting front
n = porosity
d = depth of ponding
hf = water entry pressure
The cumulative infiltration is simply I = nL.
To use this equation you must find the values of Ksat and
hf which give the best fit to the data.
35
Permeability: Tension infiltrometer
Direct Reading Scale
Applies water at set
tension via Marriotte
bottle
Using at sequence of
pressures can get K(h)
curve
Read flux using pressure
sensors
Introduced in 1980’s,
becoming the industry
standard
Marriotte
Bottle
Septum
Filling Port
36
Interpreting Tension Infiltrometer Data
The data from the tension infiltrometer is
typically interpreted using the results for steady
infiltration from a disk source develped by
Wooding in 1968 for a Gardner conductivity
function K=Ksexp(-t)

2
Q = r Ks exp( h) 1 +

4 
[2.169]

 r 
r is the disk radius. Using either multiple
tensions or multiple radii, you can solve for Ks
and 
37
Typical Tension infiltrometer Data
BOREAS 1994
Tension Infiltration Test
NSA-YJP
8 cm Disk
DOY 251 (c)
8.00
Depth Infiltrated (cm)
7.00
6.00
5.00
3 cm
4.00
3.00
6 cm
2.00
15 cm
1.00
0.00
0
400
800
1200
1600
2000
Time (s)
38
BOREAS 1994
3 cm Tension
Interpretation requires
fitting a straight line to
the “steady-state” data.
Note: noise increases
as flow decreases
y = 0.0032x + 0.6687
R2 = 0.9982
9.00
Depth Infiltrated (cm)
8.00
7.00
6.00
5.00
4.00
3.00
2.00
1.00
0.00
0
400
800
1200
1600
2000
Time (s)
BOREAS 1994
6 cm Tension
BOREAS 1994
15 cm Tension
1.00
y = 0.0003x + 0.1412
R2 = 0.9441
0.90
2.50
Depth Infiltrated (cm)
Depth Infiltrated (cm)
3.00
y = 0.0009x + 0.575
R2 = 0.9672
2.00
1.50
1.00
0.50
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
0.00
0
400
800
1200
Time (s)
1600
2000
0
400
800
1200
Time (s)
1600
2000
Permeability: Well permeameter
Establishes a
fixed height of
ponding
Graduated Cylinder
Bubbler
Shut of f valve
Support plate
Measure rate of
infiltration
Can estimate
K(h)
relationship via
time rate of
infiltration
Device Outlet
H
a
40
Making sense of Well Permeameter data
Interpretation of well permeameter data typically
employs the result of Glover (as found in Zanger,
1953) for steady infioltration from a source of radius a
and ponding height H Q = 2 H2 K /C
[2.173]
fs
The geometric factor c is given, for H/a<2 by
C = sinh
-1
(H/a) - [(a/H)
2
+ 1] 1/ 2 + a/H
[2.174]
For H/a>2, error can be reduced by using Reynolds
and Elricks result
CQ
Kfs =
[2.175]
2
2
*
2 H + a C + 2 H/ 
Where * is tabulated
41
Ks - Lab methods: constant head
Basically reproduces
Darcy’s experiment
Constant
Head in flow
h2
Important to measure
head loss in the media
Typically use “Tempe
Cells” for holding
cores, which are widely
available
Packed
Column
h1
Constant
Head out flow
42
Ks - Lab methods: falling head
Better for low permeability
samples.
Need to account for head
loss through instrument
Measure time rate of falling
head and fit to analytical
solution
r x  h1 
k  2 Ln 
R t  h2 
radius r
Core
radius R
2
43
Measuring Green and Ampt Parameters
The Green and Ampt
infiltration model
requires a wetting
front potential and
saturated
conductivity. The
Bouwer infiltrometer
provides these
parameters
[WRR 4(2):729-738,
1966]
2r
Volum etric ly
Gra duate d
Re servoir
W ater Flood Valve
Vacuum
Gauge
Air Purge Valv e
H
O-Ring Seal
G
SteelW ool
2
R
L
Approx
10 cm
44
2r
The Device
Volum etric ly
Gra duate d
Re servoir
Key Parts:
Reservoir
Pressure Gauge
Infiltration Ring
W ater Flood Valve
Vacuum
Gauge
Air Purge Valv e
H
O-Ring Seal
G
SteelW ool
2
R
L
Approx
10 cm
45
Identify the Air and Water Entry Pressures
ha – air entry pressure
hw – water entry pressure
(1)
(5)
Typically assume that
(6)
ng curve
ni
an
sc
y
ar
nd
co
se
e
rv
cu
i ng
ann
primary sc
(8)
ha = 2 hw
ha
(2)
hw
ma i
n
drai
ning
curv
e
(4)
(7)
ma in wett ing
curve
(3)
0
0
r

46
su
s
Procedure
1. Pound Ring in with slide hammer about 10
cm
2. Purge air and allow infiltration until wetting
front is at 10 cm
3. Measure dH/dt to obtain infiltration rate
4. Close water supply valve
5. Record pressure on vacuum gauge: record
minimum value
47
Employ falling head method for Ks
Recall standard falling head result
from lab methods:
r x  h1 
K fs  2 Ln 
R t  h2 
2
Remember that Kfs is about 0.5 Ks
48
Water Entry Pressure
The water entry pressure will be
taken as half the value of the
measured air entry pressure (the
minimum pressure from the vacuum
gauge on the infiltrometer)
WATCH OUT: correct observed
pressure for water column height in
unit
49
Limitations on Bouwer Method
1. All parameters are “operational” rather than
fundamental
2. Conductivity is less than K found in labs due
to trapped air
3. Rocks and cracks can render measured
value of hw incorrect.
For more details on method see:
Topp and Binns 1976 Can. J. Soil Sci 56:139-147
Aldabagh and Beer, 1971 TASAE 14:29-31
50