GEOS 4310/5310 Lecture Notes: Unsaturated

Flow

Dr. T. Brikowski

Fall 2013

0
Vers. 1.27, printed November 19, 2013
Introduction
I increasingly important in hydrology because it is the link
between the surface and the water table
I also because almost all solid waste disposal takes place above
the water table (e.g. Yucca Mountain Nuclear Waste
Repository)
I involves any water found between the ground surface and top
of the capillary fringe
I unsaturated flow is also known as partially-saturated or vadose
zone flow
I unsaturated zone pores are partially filled with water, i.e.
pores contain water + air; consequently water surface tension
plays an important role in fluid movement (Fig. 1)
I more difficult to study and model than saturated flow, only
recently addressed in great detail
I good general online reference is Natural Resources
Conservation Service Soil Survey Manual
Soil Makeup

Figure 1: Air-water-particle relationships in soil (Fig. 17.7, Keller, 2012).

Changes in the distribution of air and water generally control soil
behavior.
Unsaturated vs. Saturated Flow Regimes

Table 1: Comparison of saturated and unsaturated flow regimes Freeze

and Cherry (after 1979, , p. 44). Symbols: moisture content (volumetric)
θ, porosity φ, pressure head ψ, Pgage gage pressure (relative to
atmospheric, Pgage = 0 at the water table), air-entry pressure Pa , and
hydraulic conductivity K .

Saturated Unsaturated
Below water table Above water table (and capil-
lary fringe)
θ=φ θ<φ
P > Pa P < Pa
Pgage > 0 Pgage < 0
ψ>0 ψ<0
h measured with piezometer h measured with tensiometer
K 6= f (θ) K = K (θ)
Basic Soil Properties

Table 2: Soil volumetric and weight properties, after Keller (Fig. 3A,
2000).

Volume Weight
Va
air volume Air Wa
V Vv Vw
total volume void volume water volume Water Ww W
Vs
solid volume Solid Ws
Vv
And porosity n or φ = V
Vw
volumetric water content θv =
V 
W −Ws
ρw
= V
Vw
saturation Rs or S = Vv
Surface Tension

I arises because liquid molecules at the air-water interface are

preferentially drawn away from the air by liquid-liquid bonds
I causes the air-water interface to be curved
I curves such that the surface area between the two phase is
minimized (see also Hillel, 1980)
I similar tension arises along the interface between water and
solid
I Minimum surface energy (tension) is achieved by balancing
these two components
Capillary Rise
I study using capillary tube example (Fig. 2)
I solid-liquid interface
I against glass (polar surface), water forms an acute angle (the
wetting angle, where water is the wetting phase)
I such an angle is sought by the water all around the water-tube
interface
I tries to form the water surface in the tube into a
downward-pointing cone
I liquid-gas interface
I air-water interface also seeks to be minimized (i.e. to be
rounded) in center of tube
I draws the interface in the center of the tube upward
I lowers the water pressure in the tube, and in general water
pressure inside a capillary tube is lower than atmospheric, by
an amount equal to the total surface-tension
I non-wetting fluids work the opposite way, hence don’t trap air
in the pores (e.g. Wikipedia)
Capillary Tube

Figure 2: Capillary rise in a tube Fetter (Fig. 6.1, 2001).

Capillary Fringe

I Knowing the tube radius, surface tension and the wetting

angle, the capillary rise (or fluid pressure) can be calculated
(eqn. 6-15, Fetter, 2001)
I soils can be modeled as bundles of capillary tubes
I a capillary fringe forms above the water table, where pores are
saturated but fluid pressure is negative
Soil Moisture Balance

I as in the saturated zone, a water (moisture) mass balance can

be performed for the soil (Fig. 3–4)
I this is an important activity in agriculture, rangeland and
forest management
I Definitions:
I field capacity of soil: minimum soil moisture content resulting
from pure gravity drainage (Fig. 6.5, Fetter, 2001)
I wilting point: minimum soil moisture content produced by
gravity drainage + plant evapotranspiration. Always lower
than field capacity (Fig. 5)
I groundwater recharge cannot occur unless soil moisture
content exceeds field capacity
Moisture Budget

Figure 3: Soil moisture budget for a farm field Fetter (Fig. 6.3, 2001).
PET computed using Thornthwaite Method, actual ET computed.
Annual Moisture Variation

Figure 4: Hypothetical annual variation of soil moisture Fetter (Fig. 6.4,

2001). Note especially groundwater and soil moisture recharge periods
Texture vs. Water Content

Figure 5: Dependence of water content on grain size. Field capacity is

maximum storage possible under gravity drainage, wilting point is
minimum storage under gravity drainage only. After Fetter (Fig. 6.5,
2001).
Pressure Head and Tension

I because of capillary forces, fluid pressure (or pressure head ψ)

are generally negative (when given as gage pressures)
I soil scientists often refer to these as suction or tension head,
and omit the negative sign
I fluid pressure and soil moisture content are directly related
because of capillary forces
Moisture Content: Measurement
I laboratory: gravimetric analysis (wet then dry weighing of soil
sample)
I field:
I tensiometer (measure pressure directly, good for low-suction
settings, Fig. 6)
I resistance cells ( gypsum blocks, dissolve in ∼1 year, Fig. 7)
I thermocouple psychrometer, good for high-suction settings
(Fig. 8)
I TDR or time domain reflectometry (Fig. 9)
I Sends signal down a pair of conductors (waveguides)
I speed of wave depends on interaction with surroundings,
determined by the dielectric constant of the soil
I that varies directly with moisture content
I same principle as ground-penetrating radar (GPR)
I accurate, flexible, mildly expensive
I most widely used is the Hydraprobe
I neutron probe (Fig. 10)
I Given a source of fast (high-energy) neutrons
Moisture Content: Measurement (cont.)

I these interact with pore water producing slow (thermal)

neutrons
I measuring thermal neutron density indicates moisture content
I expensive, accurate at all suctions, but risky (neutron source
required)
Tensiometer

Figure 6: Tensiometer for measuring soil suction in moist settings. Tube

is partly filled with water, low-permeability porous cup at bottom allows
pressure equilibration with soil, pressure gauge measures suction (pore
pressure relative to atmosphere). See also SoilMoisture Inc..
Gypsum Block Moisture Sensor

Figure 7: Gypsum (electrical resistance) block for measuring in-situ

moisture content. Blocks equilibrate with soil, resistance of gypsum vs.
moisture content known, and used to quantify moisture content. Blocks
dissolve with time, leading to drift in measurements. After Hillel (Fig.
7.2, 1980). See also SoilMoisture Inc..
Thermocouple Psychrometer

Figure 8: Thermocouple psychrometer for measuring in-situ soil moisture

content. Functions by measuring relative humidity within porous ceramic
cup, which essentially indicates saturation. Effective in moist and
relatively dry settings, after Hillel (Fig. 7.13, 1980).
TDR Probe

Figure 9: Time-domain reflectometry probe. Signal propagates down

probes, reflects from end, travel time is measured at top unit. See
SoilMoisture, Inc. and TDR animation.
Neutron Probe

Figure 10: Neutron probe for measuring soil moisture, especially useful
for vertical profiles in boreholes. After Hillel (Fig. 7.3, 1980).
Characteristic Curves

I the relationships between pressure head ψ, volumetric

moisture content, and hydraulic conductivity, are typically
described graphically via characteristic curves (Fig. 11).
I Moisture Retention Curve (Fig. 13)
I moisture content θ can be measured for a soil held under
varying pressure head ψ
I this is plotted as a moisture retention curve (e.g. Fig. 6.7,
Fetter, 2001).
I two curves are generally found, depending on whether the
sample is being progressively dried or wetted. This is hysteresis
I air-entry value:
I as a sample is gravity-drained from an initially saturated state,
saturation changes little at first
I eventually the saturation begins to change rapidly with little
change in ψ, as pores begin emptying, and air becomes a
continuous phase in the pores.
Characteristic Curves (cont.)
I this occurs at the air-entry value ψa of pressure head, and the
corresponding fluid pressure is the air-entry or bubbling
pressure (Fig. 13)
I Hysteresis
I hysteresis generally means a multi-valued function (e.g.
multiple values of θ are found for a given ψ, depending on the
history of the sample)
I this phenomenon considerably complicates modeling of
unsaturated flow
I Conductivity curve
I measured by fixing the sample θ and observing the water flow
rate. Darcy’s Law is used to obtain K (θ)
I conductivity decreases as pressure declines (i.e. with
decreasing saturation) because the water phase in the pores
becomes poorly connected.
I when log K is plotted against ψ, the curves are quite similar to
those for θ vs ψ (Fig. 14) or (e.g. Fig. 6.8 Fetter, 2001)
Characteristic Curves (cont.)

I large pores will drain more quickly, which can preferentially

direct flow (e.g. in the matrix rather than fractures in
fractured rock), or lead to surprising results (clay often has
higher conductivity than sand or gravel at low water content)
(Fig. 15)
Example Characteristic Curves

Figure 11: Idealized curves of tension head (ψ), hydraulic conductivity

(K) and water content (θ) After Fetter (Fig. 6.8, 2001).
Moisture Retention vs. Grain Size

Figure 12: Available moisture vs. grain size, after soil saturation
summary. “pF” is the logarithm of the height of water column (cm) that
would yield the necessary suction.
Water Content vs. P, Clay Loam

−300

drying
Ψ
−200
(cm)

Ψa
−100 wetting

0
θ sat = 0.32
0.10 0.20 0.30

θ
Figure 13: Schematic moisture retention curve showing typical relationship between pressure head ψ and
moisture content θ in unsaturated materials. Hysteresis shown, where wetting curve always lies to the higher ψ
side of the drying curve. See also Fetter (Fig. 6.8, 2001) and Freeze and Cherry (Fig. 2.13, 1979).
K vs. Water Content, Clay Loam
Clay Loam Characteristic Curve
10

0.1
K(Θ) (cm/hr)

0.01

0.001

0.0001

1e-05
5 10 15 20 25 30 35 40 45
Moisture Content Θ (%)

Figure 14: Schematic unsaturated conductivity curve showing typical

relationship between moisture content θ and hydraulic conductivity (K)
in unsaturated materials. See also Fetter (Fig. 6.8, 2001) and Freeze and
Cherry (Fig. 2.13, 1979).
K vs. Grain Size

Figure 15: Dependence of hydraulic conductivity on water content and

grain size. Note crossover where coarse sediments become more
permeable at high water contents. After Fetter (Fig. 6.10, 2001).
Unsaturated Flow Equation

I head in the unsaturated zone is the sum of the elevation z

and pressure head ψ, just as in the saturated zone
I the flow equation is different than the saturated case, since
conductivity is a function of ψ or θ
I Flow Equation (AKA Richards Equation, the form below
assumes ∇ψ  1)

∂θ
= ∇ · (K (ψ)∇ψ)
∂t  
∂ ∂ψ
= K (ψ) + ···
∂x ∂x
I solution of this non-linear equation can be difficult, but a
number of good numerical models are available (e.g.
HYDRUS2D)
Storm Infiltration Pulse

Figure 16: Downward movement of infiltration pulse in a soil, with

porosity of 29% (i.e. θ = 0.29 is full saturation). Field capacity is 0.06.
After Fetter (Fig. 6.9, 2001). See also resistivity results from Norway
airport report.
Drought Pulse

(a) 0.5m (b) 1.5m

(c) 1.0m (d) 2.0m

Figure 17: Soil moisture variation at 1 m depth, El Reno, OK 1998-2007. Note propogation and persistence
of moisture deficit with depth. From Garbrecht, Schneider, and Brown (2007).
Texas Vertisol Summary

I Central Texas (primarily along the Edwards-Trinity groups) is

characterized by Vertisols, a taxonomic soil group that
exhibits shrink-swell behavior (Fig. 18)
I one well-studied site is Riesel, TX (USDA-ARS Grassland, Soil
and Water Research Laboratory, Allen et al., 2005)
Vertisols in Texas

Figure 18: Distribution of soils with high shrink-swell potential in Texas

(Vertisols). After USDA.
Shrink-Swell in Soils

Figure 19: Shrink-swell process in Vertisols, and its consequences (Fig.

17.11a-b, Keller, 2012)
Dallas Sat-Unsat Zone Interaction

Figure 20: Seasonal changes in aquifer-soil interaction in Blackland

Prairie soil (Vertisol, Fig. 11, Allen et al., 2005).
Dallas Sat-Unsat Zone Interaction 2

Figure 21: Seasonal changes in rainfall-runoff in Blackland Prairie soil

(Fig. 13, Allen et al., 2005). This study also found no runoff occurs
until 20% of annual rainfall has re-moistened the soil.
DFW Local Runoff Curve
7 0.22

Enhanced
0.2
Runoff Coefficient
6 Drying

0.18

Monthly Runoff Coefficient

5 0.16
Monthly Precip (in)

0.14

%
4

25
r>
0.12

AP
JJ
3 0.1

0.08
2 Normal Precip
0.06

1 0.04
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month

Figure 22: DFW climate normals and high-JJA (summertime) average

precipitation . Monthly runoff coefficient is computed as the ratio of
monthly normal runoff divided by monthly normal precipitation for the
basin.
DFW Local Runoff Curve (cont.)

I As water stress increases, significant changes in agricultural

and urban irrigation will be required.
I One method, surge irrigation (Fig. 23), enhances crop water
efficiency significantly (Fig. 24).
I a casualty of these issues is rice farming in Australia
Surge Irrigation

Figure 23: Surge irrigation principles, after U. Neb. Left shows

alternating distribution of irrigation surges, right shows resulting soil
moisture infiltration.
Surge Efficiency

Figure 24: Surge irrigation benefits: increased irrigation efficiency (water

productivity, or increased crop yield per unit of water). From a study of
Vertisol agriculture, Ethiopia (Jiru and Van Ranst, 2010).
References
Allen, PM et al. (2005). “Field data and flow system response in clay (vertisol)
shale terrain, north central Texas, USA”. English. In: Hydrological
Processes 19.14, pp. 2719–2736. issn: 0885-6087. doi: 10.1002/hyp.5782.
Fetter, C. W. (2001). Applied Hydrogeology. 4th. Upper Saddle River, NJ:
Prentice Hall, p. 598. isbn: 0-13-088239-9. url: http://vig.prenhall.
com/catalog/academic/product/0,1144,0130882399,00.html.
Freeze, R. A. and J. A. Cherry (1979). Groundwater. Englewood Cliffs, NJ:
Prentice-Hall, p. 604.
Garbrecht, J. D., J. M. Schneider, and Glenn. O. Brown (2007). “Soil Water
Signature Of The 2005-2006 Drought Under Tallgrass Prairie At Fort Reno,
Oklahoma”. In: Proceedings of Oklahoma Academy of Science 87,
pp. 37–44. url: http://www.ars.usda.gov/SP2UserFiles/Place/
62180520/2007_OSA_87()37-44_ARS.pdf.
Hillel, D. (1980). Applications of soil physics. Academic Press, New York,
p. 385. isbn: 0-12-348580-0.
Jiru, Mintesinot and Eric Van Ranst (2010). “Increasing water productivity on
Vertisols: implications for environmental sustainability”. English. In: Journal
Of The Science Of Food And Agriculture 90.13, pp. 2276–2281. issn:
0022-5142. doi: 10.1002/jsfa.4082.
References (cont.)
Jury, W. A., W. R. Gardner, and W. H. Gardner (1991). Soil Physics. ISBN
0-471-83108-5. New York: John Wiley and Sons, p. 328.
Keller, E. A. (2000). Environmental Geology. 8th. Upper Saddle River, NJ:
Prentice Hall. isbn: 0-13-022466-9.
– (2012). Introduction to Environmental Geology. 5th. Prentice Hall. isbn:
9780321727510. url:
http://www.pearsonhighered.com/educator/product/Introduction-
to-Environmental-Geology-5E/9780321727510.page.
Simunek, J., M. T. van Genuchten, and M. Sejna (1999). “Using the
HYDRUS-1D and HYDRUS-2D codes for estimating unsaturated soil
hydraulic and solute transport parameters”. In: Proceedings of the
international workshop on Characterization and measurement of the
hydraulic properties of unsaturated porous media. Ed. by
M. T. van Genuchten, F. J. Leij, and L. Wu. Monograph. U.S. Dept. of
Agriculture, U. S. Salinity Laboratory. Riverside, CA: U. California-Riverside,
pp. 1523–36.
Tindall, J. A. and J. R. Kunkel (1999). Unsaturated Zone Hydrology for
Scientists and Engineers. Upper Saddle River, N.J.: Prentice-Hall. isbn:
0-13-660713-6.