Sedimentation and Remote Sensing

opossumoozeMechanics

Feb 21, 2014 (3 years and 6 months ago)

79 views

Sedimentation and Remote Sensing



Introduction: A certain amount of released earth materials into water or the atmosphere
is a natural occurrence; however, excessive sedimentation is of concern when
environmental or commercial problems arise from the pr
ocess. It is also of concern when
the sedimentation is clearly the result of
erosion,
possibly due to
urbanization, poor
farming practices and industries such as mining.


The tie to physical science is in the physics of soil processes such as infiltrati
on, runoff
and permeability.
Infiltration is the rate of entry of precipitation into the surface of soil,
runoff is the percentage of precipitation which does not enter the soil and permeability is
the ease with which water can travel
through the soil pr
ofile, usually measured as a rate
as well
.



Infiltration rate is dependent upon several factor
s which are difficult to quantify,
including soil type (textural)
, soil structure and slope.
There is a dual nature to soil on a
gradient and all of it is inde
ed “on a slippery slope.” If infiltration is not sufficient,
runoff occurs and pulls soil particles with it in a steady stream. However, if infiltration is
too great, massive earth movement called mudslides can occur.




http://www.ent.iastate.edu/imagegal/practices/tillage/conventional/erosion.html
U.S.G.S. Pu
blic Affairs Office, Menlo Park, CA.
U.S.G.S Public
Affairs Office Menlo Park, California


Infiltration rate is lowest for clay par
ticles in a massive structure on
a steep slope.
Clay

profiles are characterized by a larger volume of pore space than
sand
, b
ut all pores are
much smaller than th
ose for sand. Movement
through the soil is not due to a
gravitational gradient (as in sand) but is due to electrostatic attraction between hydrogen
in water and electronegative elements in the soil (oxygen, silicon, et
c.) The overall
process is slow and there is usually not sufficient time
for absorption
due to rapid water
flow on a steep grade unless there is sufficient vegetation to trap water for a longer time
period.


Clay or sand are soil textures
; soil structur
e consists of a secondary organization of soil
particles into “shapes” such as granular, cubic, columnar or one large mass
(massive)
.
The advantage of cubic or columnar
shapes
for infiltration is that channels exist in the
soil which are larger than the i
ndividual pores
and allow more rapid infiltration.





http://nesoil.com/gloss.htm

blocky structure of subsoil



http://www.evsc.virginia.edu/~alm7d/soils
/images/images1.html




There are various equations for fluid flux through a permeable material; Darcy’s law is
probably best known:

q =
-
k (Pb

Pa)/


w
here q is fluid flux, k is permeability, Pb is pressure at base of fluid front, Pa is pressure
at f
luid head and

is viscosity of the fluid.
Flux is measured as m
3
/m
2
/s, which reduces
to m/s.




There are othe
r equations which consider additional
criteria, but all are dependent on
permeability. One equation for hydraulic conductivity, which is
simi
lar to Darcy’s is:

K = k
γ
/



w
here K is hydraulic conductivity,
k is permeability
,
γ
is specific weight of water
, and


is viscosity of water.
For “standard” conditions, hydraulic conductivity is very c
lose to
permeability. Hydraulic conductivity te
chnically can vary
with conditions
of flow
while
permeability is a property of the soil itself.


Permeability can be calculated with an equation:

k = C d
2
where k is permeability, C is a configuration constant and d is average pore
diameter. Configu
ration constants can be estimated from texture and structure, but
usually permeability is best measured
empirically.



Table 1. Size limits (diameter in millimeters) of soil separates in the USDA soil textural classification system.

Name of soil separate



Diameter limits (mm)



Very coarse sand*

2.00
-
1.00

Coarse sand

1.00
-
0.50

Medium sand

0.50
-
0.25

Fine sand

0.25
-
0.10

Very fine sand

0.10
-
0.05

Silt

0.05
-
0.002

Clay

less than 0.002

* Note that the sand separate is split into
five sizes (very coarse sand, coarse sand, etc.). The size range for sands,
considered broadly, comprises the entire range from very coarse sand to very fine sand, i.e., 2.00
-
0.05 mm.

edis.ifas.ufl.edu
/SS169



We are going to equate infiltration with pe
rmeability although they are technically
different. Infiltration is dependent on permeability but also
soil
surface conditions.
Permeability is typically measured under saturated flow con
ditions, which we will
emulate in the laboratory, but infiltration
rate can vary widely due to incoming rate of
water.


In the following laboratory, we will measure permeability for two soils
-
a clay and a

sand
-
and look at
permeability for
these
soils on a slope.




Laboratory procedure:


1.

Set up a canister with holes
on the bottom, lined with filter paper (thin) and fill
with sand
about the one
-
fourth

of the canister height
.

2.

Position
the canister on a
rings
tand or other upright apparatus and clamp a hose or
buret above the canister. A ring with a wire gauze between
the hose and soil will
help disperse the water over the soil.

3.

Place another empty canister (without holes) below the soil canister
(diagram A).

4.

Slowly saturate the soil, then o
ne student must
adjust the
faucet or buret
flow until
the rate produces ponding
water and then back off to a flow where no ponding
occurs.


5.

Measure the
leached
water
height in cm after about ten minutes and then divide
this value by 10 to get
permeability
flow rate in cm/min.







6.

For soil on a slope:
Remember the fauce
t speed used in setup A and use this in
setup B. The only difference here is that the
soil canister
w
ill be put on a slope of
about 1
0
degrees.

7.

Repeat the same procedure as for setup A (at the same flow rate) and
collect
leached water in the lower caniste
r for 10 minutes. Calculate permeability in
cm/min.

8.

Note rate
of runoff as well.
After 10 minutes, collect the water accumulated on
the downward side of the surface with a pipet and place this into a canister of the
same size as the others and note heigh
t in cm. Divide this by 10 to get runoff rate
in cm/min. (
If there is no runoff, increase the steepness
of the soil until there is
runoff
and run the experiment again with measurements.
Note the angle of
inclination in your notebook.
)

9.

Repeat the entire
process for a clay soil.
If there is time, repeat the process for
clay with plants “planted” in the soil.





A slightly loose hose dispersed water well
.
A buret was somewhat more precise in

pinpointing




necessary
flow
rate.




Using a protractor to measure slope

P
ipetting
runoff from soil at 10
-
degree incline








Questions:



1.

List permeability
for the following:

Sand, flat:
0.1 cm/min

Sand, tilted:

0.06 cm/min

Clay, flat:

0.02 cm/min

Clay, tilted:

0.005 cm/min

Clay, tilted, with vegetation:


2.

The ra
te of flow for the flat soils actually represents infiltration rate which is just
below the ponding rate. Any type of ponding is considered to be potential runoff
and erosiv
e even when on a “flat” surface,
due to imperfections in terrain.

Whic
h soil tolerated a higher
rate of “precipitation” without ponding
?


sand


3.

For the same angle, which type of soil produced more runoff, sand or clay?


clay


4.

What was the nature of the runoff water? (Did it contain soil, etc.)

Containe
d small particles of soil


5.

See if the following relationship tentatively worked out in class is a good
predictor of runoff rate in cm/min:


sin
Θ

x

(
permeability rate on
0
-
degree slope
)


Where:

Θ
is the angle of slope for the canister

Permeability is the rate
at 0
-
degree slope
in cm/min


For sand at no slope, permeability was 0.1 cm/min

Calculated runoff for sand at 10
-
degree slope: Sin(10) x
0.1 = 0.017 cm/min

Actual runoff rate for sand at 10
-
degree slope: 0.016 cm/min


6.


a.
Determine the gravitational
acceleration

on a discrete
particle
of water


at the top of a slope which is 14.7 m long at
an inclination of
20 degrees.

a.

Det
ermine the velocity of this particle
of water at the bottom of the
incline.

b.

How much time will it take the water to reach the bottom of the slope?




7.

a.
For a sphere of water of 0.0042 cm
3
, calculate its mass and gravitational force
it
p
os
sesses
on

this
slope.


0.0042 g;

F = 0.0042 x 3.35 = 0.014 N


a.

The sphere of water will only be able to move a soil sphere of equal size
or smaller (due to contact). Assume a coefficient of friction for the
soil
sphere of 0.
9
and a density of 2.65 g/cm
3
. What is the radius of the largest
sphere the water will move? What classification is it? (sand, silt, clay).



Assume the gravitational force Fp
of the water on this slope
is translated
into the lesser horizontal force
Fh = Fp (cos 20)
once it reache
s the
bottom:

Fp = 0.014 N

Fh = 0.013 N


Assume Fh is equal to frictional force Ff to produce movement of constant
velocity (not acceleration). This corresponds to the largest soil particle
which can be moved.

Ff
=

Fn

0.013 = 0.9
Fn
where Fn is the weight of the soil particle

Fn = 0.0144
N

The mass of the particle =
0.0144 N/9.8 =
0.
00147
g

2.65 g/cm
3
= 0.00147
g/x x =
0.000556 cm
3


Vol of sphere = 0.000556 = 4/3
π
r
3
r =
5.1 x 10
-
2
cm
;

Medium sand



Relating sedimentat
ion to remote sensing:


Sedimentation can benefit agriculture by depositing nutrients on flood plains and
extending delta land, but also costs humans in terms of flood damage, waterway
clogging, poor water quality, and recreational site damage. Of late
it is of increased
concern due to effects on environments which are fragile: estuaries, wetlands
, coral reefs

and continental shelves.


Erosion is increased soil loss and sedimentation due to poor supervision of human
activities. The main causes of erosi
on include lack of vegetation on
agricultural land,
overgrazing, deforestation and
mining operations.


View the following satellite images and
see if you can identify the location
and find the
sedimentation
source
:



Image of the Ganges River delta and
the Bay of Bengal acquired by the Moderate Resolution Imaging Spectrora
diometer (MODIS). This image
shows the massive amount of sediments delivered to the Bay of Bengal by the Ganges River, sediments
that are derived from erosion of the Himalayan mountain range to the north. Click on this image to see a
large high
-
resolution
version that includes the Himalayan range. Mt. Everest, the highest point in the
world, is located in the upper right corner of the high
-
resolution image.

http://daac.gsfc.nasa.gov/oceancolor/scifocus/oceanColor/sedimentia.shtml



SeaWiFS image of the U.S.
East Coast acquired one week after the passage of Hurricane Floyd (see image
below). The sediments generated by the flood waters of rivers in Nort
h Carolina are
seen entering the Gulf Stream off of Cape Hatteras. Also note the increased
turbidity in the sounds and river estuaries and persistent sediment suspension
southward along the coast.
http://daac.gsfc.nasa.gov/oceancolor/scifocus/oceanColor/sedimentia.shtml



SeaWiFS image of Italy and the
Adriatic Sea. The Balkans to the west and the snow
-
covered Alps to the north are also visible. The Po
River valley is the hazy brown area just south of the Alps. The plume of s
ediments carried by the Po River
is seen on the western side of the far northern Adriatic Sea.
http://daac.gsfc.nasa.gov/oceancolor/scifocus/oceanColor/sedimentia.shtml


A large sediment plume enters the
Mozambique Channel south of the resort town of Beira.
(Satellite photo courtesy
NASA
)



http://www.star.le.ac.uk/edu/Probes.shtml

Conjectured water channels on the red planet.


Here are a few interesting pictures of wind erosion as well:
See if you can identify th
e
location and the extent of wind
-
blown debris.

http://www.msmedia.homestea
d.com










Assignment:

1.

Find erosion statistics for North Dakota: How much soil is lost per year by water erosion and by
wind eros
ion?
What is the tolerable limit set forth by the USDA?


Source: USDA
-
NRCS



Most sources list 5 tons/acre per year as the tolerable limit.


2.

Find satellite images of sedimentation in rivers in the Midwest (North Dakota if possible.) T
he
1997 Red River flood might be a good case study, if satellite images are available.


The following images show Red River flooding in 1997; although only water levels are shown, the amount
of sedimentation from the event can be surmised.



March 1997 before flood

Red i
s snow cover; yellow is cloud cover

April 1997 during flood


May 1997 after flood





www.math.montana.edu/.../rrf/flood_pics.html