HydroPedology New Solution to an Old Problem

1
HydroPedologyNew Solution to an Old Problem

• Rabi H. Mohtar

Purdue University, Agricultural and Biological
Engineering
Presentation to the MAE Center University of
Illinois October 3, 2006
2
Pedotransfer Functions (PTFs)
Takes what you have and gives you what you need!
They are of great use and got us where we are and
will always be used as long as there is limited
data. However, by in large they are empirical and
do not allow for scaling of processes.
Hillel, 1980 (Introduction to Soil Physics,
Academic Press, Inc.)
PTF
3
Soil Water Models

• Challenges
• Parameterization of the soil medium is mostly
  empirical and ignores the interaction between
  soil water and soil structure
• Models simulate rigid medium and do not reflect
  the hydrostructural dynamic of the medium as it
  swells and shrinks
• Linkage between micro and macro domains is not
  explicit/dynamic
• Accordingly, models do not allow for scaling of
  processes and transferability of information
  across scales

Van Genuchten and Simunek (2004) after Altman et.
al, (1996)
4
Disconnection Between Pedology and Soil Physics
in the Three Fundamental Axes Describing Soils
Axe III Functionality
Soil Physics Uses the continuous porous media
theory. With no consideration of soil structure
the medium is a homogeneous mixture of solids,
liquid and air.
There is a disconnection between the functional
axe III and the other axes that describe the soil
morphology and evolution.
Continuity Equation
Axe I Evolution
? Volumetric Water Content (m3/m3)
Pedology Science of the soil organization in
axes I and II plane
Axe II Structured material
5
Unified Description of the Soil Medium Processes
and Fluxes
Axe III Functionality
Specific Structural Volume (dm3/kg)
To take soil structure into consideration, the
structural water content, W, and the apparent
volume, V, must appear explicitly in all
equations describing soil structure and its
behavior ? W/V.
Structural Water Content (kg/kg)
Notamment dans léquation de continuité, base de
toutes les équations de transfert
Axe I Evolution
EVOLUTION
Axe II Structured material
V and W are organizational variables including
them in the equation allows for connecting the
three axes
6
Introduction
Objectives

• The objectives of this talk is to present a
  comprehensive model of the vadoze zone structured
  soil-water medium in which the thermodynamic
  equilibrium and kinetics are characterized by its
  volume and hydro-structural changes and
  properties.
• Specifically, I will discuss
• Soil water medium characterization
• Conceptual model development
• Model application
• Delineation of Functional soil mapping Units

7
Soil Structure and Functionality
Clay particles
Primary peds
Inter-ped pore space
Mineral grains
Primary soil
mapping unit
Clay pore space
Primary soil mapping unit
Soil type
REV
Pedostructure
Primary ped
Horizon
Geomorphological unit
Clay plasma porosity (micro-porosity)
Vertical porosity (cracks, fissures)
Interpedal porosity (macro-porosity)
Pedostructure, primary peds, primary particles,
are functionally defined and quantitatively
determined using the shrinkage and potential
curve measurement
Primay particles and pedological features
Pedostructure
Primary peds and free mineral grains
8
Internal Hydro-structural Configurations of Soil
Characterization
Conceptual Model
Hydrostructural states
Each particular hydrostructural configuration of
the soil medium is characterized by both 1) a
water saturation state (dry, unsaturated,
saturated) and 2) a swelling state (minimum,
intermediate, maximum) in each of the micro and
macro-pore systems.
Braudeau, Franji, and Mohtar, SSSAJ, 2004
9
Soil-water-chemicals-air processes in soil
Conceptual Model

• This representation allows
• for the following
• Linking soil properties to soil behavior
• Inclusion of overburden pressure effect on soil
  water
• Drainage from macro flow
• Transfer of chemicals between micro and macro
  pore systems
• Dissolution
• Sorption/desorption at the surface of primary
  peds
• Easier integration into water quality modeling

Kv (volatilization)
Kdis (dissolution)
Ks (striping)

Kb(biodegradation)
Kd sorption/ desorption
Adsorbed phase
Dissolved phase macro and micro
In-gas phase
10
Four Basic Measurable Soil-Water Properties
Characterize the Pedostructure
Soil Swelling Curve soil specific volume as a
function of time, V(t)
2
Soil Shrinkage Curve specific volume as a
function of water content, V(W)
1
4
Unsaturated macropore Hydraulic Conductivity
Curve , kma(Wma)
Macropore and micropore Water Potential Curves,
hma(Wma) and hmi(Wmi)
3
11
The complete set of the pedostructure
(hydro-structural) parameters
Characterization

• They are a total of 14 independent parameters
  needed to characterize and describe the water and
  pedostructure interaction

Braudeau and Mohtar, SSSAJ, 2006
12
Principles for the Pedostructure Parameters
Estimation (KamelSoil)
Characterization

• Determination of 9 out of 14 parameters from soil
  database
• Using the best fit of simulated and measured
  water potential and conductivity curves

Soil data bases Texture, Soil Org.
Matter Field capacity Wilting point
Braudeau and Mohtar, SSSAJ, 2006
13
Principles for the Pedostructure Parameters
Estimation (KamelSoil)
Characterization

• The rest of the needed parameters, namely, WN,
  kN, Kbs, kmi, and Emi, can be estimated according
  to the following procedure

Braudeau and Mohtar, SSSAJ, 2006
14
Functional Model of the Pedostructure
Conceptual Model
The pedostructure is the Representative
Elementary Volume (REM) of the soil fabric.
Arrows represent the three types of controlled
water flux. Vp means poral volume and mS is the
mass of the structured solid phase contained in
the REV. In this REM, three distinct flows are
represented 1 macro to macro 2 micro to micro
and 3 micro_macro transfer. 1 and 2 are
Darcian flows governed by potential gradient at
equilibrium states (shrinkage curve), while 3 is
governed by the swelling process of soil primary
peds on its way to equilibrium. (Braudeau and
Mohtar 2004 and 2006
Braudeau and Mohtar, SSSAJ, 2006
15
Model Inputs

• 1. Soil fabric (pedostructure) characteristics
• The four curves presented, continuously
  measured, or their parameters approximated using
  PTF. These parameters can be studied as they
  evolve with time naturally or under land-use and
  management. The ideal and approximate
  characterizations are already developed and are
  being tested.
• 2. Soil surface structure characteristics
• This includes surface crusts and in-situ surface
  fissures and cracks.
• 3. Specific horizons parameters
• depth, hydrostructural properties and their
  evoluation

Kamel
16
Model Outputs

• Hydrostructural characteristics (state variables)
• Bulk density at dry and water saturated states
• Hydraulic Conductivity for micro and macropore
  spaces
• Available water reserve
• Field capacity wilting point drainage water
  residual water
• Micro and macro porosity
• COLE index
• Additional outputs (rate variables)
• Spatial and temporal distribution of soil
  moisture and air in micro macro systems.
• Spatially distributed fluxes between micro-macro
  exchange
• Spatially distributed fluxes between layers and
  horizons
• Spatial and temporal water distribution of all
  water reserves

Kamel
17
Soil Medium Discretization

• Soil Surface

• Layers

• Horizon A1

In this representation the layers move in the
vertical space due to volume change of the soil
fabric. However, they always maintain the same
structural solid mass. The hierarchy of pedon,
horizon, layer, pedostructure and primary peds is
conserved.

• Other Fabric

• Horizon A2

• Gravels

• Layer thickness
• ?z HhrzBi/nhrzBi
• Layer depth
• z ??z above

• Horizon B1

• HhrzB1

• Horizon B2

Kamel
18
Kamel Main Screen Input Parameters Window
Kamel
19
Kamel Execution Window
Kamel
20
Kamel Output Screen

• Output of W and Wmi with execution time (no. of
  time step) for a specific depth (10 cm) using
  plotter tool.

Output of Wma with time. Blue line is at 10 cm
depth, red line at 20 cm depth, green line at 30
cm depth
Kamel
21
Soil Moisture Profiles from Kamel Output
Compared to Experimental Data
W0J initial moisture experimental data W1J
measured moisture profile after 1 day W60J
measured moisture profile after 60 days. Lines
represent computed profile using Kamel.
22
Kamel Sample Simulation Results

• Hysterisis effect and the role of micro and macro
  transfer.

Kamel
23
Comparison of FC and PWP values calculated using
nine PTFs with those calculated using the PS Model
Estimation of field capacity and wilting point
0.5
Estimation Method
Field capacity
0.45
Wilting point
0.4
Baumer
Brakensiek
0.35
BSS SubSoil
0.3
BSS TopSoil
Campbell
0.25
Water content m3/m3
Epic
Hutson
0.2
MANRIQUE
0.15
RAWLS
CR (2)
0.1
CR (1)
0.05
0
calcaire
salé
vertique
modal
calcaire
salé
vertique
modal
Soils from the vallée de la Majerda

Braudeau, Mohtar, and Chahinian, (Elsevier, 2004)
Model Evaluation
24
ScalingDelineation of Functional Soil Mapping
Units

• The study area, landform map, and SSURGO soil map
  were overlaid following the hierarchical Systems
  Theory approach. Such that the smaller data layer
  is overlaid over the larger data layer without
  allowing for any intersecting areas. Each layer
  should be totally included in the upper layer,
  and at the same time, totally includes the lower
  layer.

25
Aerial Photo of Study Area
26
Delineation of the Functional Soil Mapping Units
and location of the Soil Samples
27
Evaluation of the Mapping Units PS surface
parameters using Discriminate Analysis
28
Uniqueness of the characterization
Evaluation of Characteristic Parameters
Discriminant analysis of the horizons of the
ferrallitic and the ferruginous soils
characterized by their pedohydral parameters.
Numbers 1, 2, 3, and 4, indicate horizons A, AB,
B1, and B2, respectively.
Braudeau, Sene, and Mohtar, EJSSS, 2004
29
Conclusions

• A methodology for the transfer of scale from
  laboratory characterization to field soil water
  modelling is presented using Representative
  Structural Volume (RSV).
• A computer model Kamel was developed based on
  this theory that addresses the scaling problem
  among measurements in laboratory, estimation from
  soil databases, and modelling at the field scale.
  The model
• Represents the soil organizational
  characteristics and variables for each
  hydrostructural state.
• Integrates between the internal physical state
  variables (structural mass of solids at the
  processes local scale) and the volumetric
  averaged large scales variables.

30
Current and Future Work

• Evaluation of Kamel
• Scaling
• Inclusion of fissures, cracks and preferential
  flow
• Modeling crust or surface layer represented by
  pedostructure parameters as influence by soil
  management
• Integration with other biophysical and
  hierarchical models

• Developing Soil information System that defines
  soil mapping units and their appropriate
  attributes including pedostructure parameters
• Integrate the characterization and modeling with
  remote sensing/observation

31
Kamel Team and Acknowledgments

• Acknowledgments
• French Embassy in USA
• IRD
• CIRAD
• Purdue University

• Team
• Rabi H. Mohtar, Purdue University
• Erik Braudeau, IRD, France
• Pierre Martin, CIRAD, France
• Pascal Clouvel, CIRAD, France
• Mohammad Salahat, Purdue University, ABE
• Majdi Abou Najim , Purdue University, ABE
• Matthieu Ronin, Rennes, France
• Carly Day , Purdue University, ABE
• Joe Mallory , Purdue University, ABE
• Adam Conklin, Purdue University
• Carol Sikler, Purdue University, ABE

32
Shrinkage curve characterizing soil structure
hierarchy
Conceptual Model
Braudeau, Franji, and Mohtar, SSSAJ, 2004
33
Properties of the Tensiometric Curve
Characterization

• Addresses the interpedal (macro-) porosity of the
  pedostructure
• Easy to measure
• A physically based equation as a function of Wma
  (gravimetric macropore water content) and
  pedohydral parameters (WM, kM, WL).
• Possibility of using the continuously measured
  tensiometric curve for determining all the
  interpedal parameters

Equation of the macropore water potential
function of Wma. Ema and s are parameters of the
tensiometric curve.
34
Pedostructure Parameters Estimation Using the
Tensiometric Curve
Characterization

• Measure the suction pressure h and the water
  content W during evaporation starting from
  saturated state.
• Make a first evaluation of (WM- s), Ema and hma
  using the excel solver to optimize the linear
  regression of h against 1/(W-WMs) over the range
  of h lt 150 mb.
• Calculate Wma as a function of W using equation 1
  where WM and kM are estimated by equations 2 in
  which s is near 0.01 and WC corresponds to the
  break-down of the tensiometer.
• Make over the whole range of W the linear
  regressions between 1) h and 1/(Wmas) and 2) of
  Wma and 1/(hhma). These Eqs. provide estimation
  of Ema, s and hma.
• Use the solver for optimizing the fit between
  calculated (Eq.3) and measured curves using as
  variable parameters WC, s, hma, Emac under
  conditions WCltWM

1
2
2
Estimated parameters
WC
35
Modeling the Conductivity Curve
Characterization

• Measure water potential at two depths, 1 and 2 cm
  below soil surface
• Measure core weight
• Using the macro water potential reading and the
  pedohydral parameters estimated in the earlier
  slide of the water potential procedure, estimate
  Wma (Eq.3), and W (Eq.1) for the two depths
• Estimate the rate of change of W1 with time
  (dm3kg-1s-1) for the layer above t2 using the top
  tensiometric readings (dm of water or kPa)
• Estimate the flux across the top surface S (dm2)
  using the weight change of the sample (kg/s)
• Calculate the flux (dm s-1) across the lower
  surface at t2 equal to the flux across the top
  surface less the rate of change of water (in dm)
  stored in the layer above t2.
• Estimate the hydraulic conductivity (dm/s) at the
  lower tensiometer t2 using Darcys law.
• Plot Wma for the lower layer against k and
  estimate the parameters of the exponential fit
  equation for the k(Wma) curve.

36
Modeling the Conductivity Curve
Characterization
kma in terms of Wma needs three
parameters kmaSat, kmaC, and a
One can demonstrate that this part of the curve
(WltWD) is exponential such as
Because of the following relationship when the
interpedal water, Wma can be considered as a film
surrounding aggregates
37
Summary and Conclusion
Summary and Conclusion

• A systematic process-based model of the soil
  medium using its volume change was developed and
  evaluated. Continuous measurements of shrinkage,
  swelling, and water potential were used to define
  and quantify the needed parameters for the model.
  The model is a physical representation of soil
  thermodynamic equilibrium and kinetics of the
  soil water medium.
• The model allows for accurate definition and
  quantification of commonly used agronomic
  properties such as field capacity, wilting point,
  air entry etc. It also allows for future
  inclusion of soil overburden, transfer of
  chemicals between soil plasma and interped
  porosity.
•  
•