“Erosion Research in Iowa” Thanos N. Papanicolaou IIHR-Hydroscience and Engineering, University of Iowa.

““Erosion Research in Iowa”Erosion Research in Iowa”

Thanos N. PapanicolaouIIHR-Hydroscience and Engineering, University of Iowa

The Clear Creek Observatory, IA

The Clear Creek Observatory, IA

The Clear Creek Watershed coordinates

HUC2 (Region)Upper Mississippi

HUC4 (Subregion)Iowa-Skunk-Wapsipinicon

HUC6 (Accounting Unit) Iowa

HUC8 (Catchment)Lower Iowa

HUC10 Clear Creek Watershed

Upper Mississippi River HUC12 North Amana SubWatershed

HUC Digit

2

4

6

8

10

12

Name

Upper Mississippi

Iowa-Skunk

Iowa

Lower Iowa

Clear Creek

North Amana

Area (km2)

491,663

58,458

32,746

4,366

267

48

Testbed(s): nested watersheds in the Upper Mississippi River Basin

The Clear Creek Observatory, IA

The Clear Creek Observatory, IA

The Clear Creek Observatory, IA

After Papanicolaou (2006), Geomorphology

Land-use Fingerprinting

The Clear Creek Observatory, IA

The Clear Creek Observatory, IA

δδ1515

N N δδ1313

CCC/NC/N

== f f (biogeochemical (biogeochemical processes)processes)

== f f ( factors ( factors ))

Using a Finnigan MAT Delta Plus isotope ratio mass spectrometer (IRMS)

The Clear Creek Observatory, IA

During decomposition, particles decrease in size. During decomposition, particles decrease in size. Also, the light N-14 atoms are incorporated into Also, the light N-14 atoms are incorporated into plant roots and thus the net change in plant roots and thus the net change in δδ1515N is an N is an increase relative to initial conditions. This is increase relative to initial conditions. This is particularly important for soil-erosion particularly important for soil-erosion considerations because particle size is a governing considerations because particle size is a governing parameter.parameter.

Decomposition, Decomposition, mechanical breaking mechanical breaking down of OM down of OM

δδ1515N= N= 2‰2‰

δδ1515N= N= 4‰4‰

Conclusions for Erosion studies:

1. The δ15N for monoculture wheat (C3) land-use reflects size distribution (after Fox and Papanicolaou JAWRA, 2007)

Results from both labs (overlapped)d 13C (%0) vs. d 15N (%0) (<53 mm)

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

-30 -29 -28 -27 -26 -25 -24 -23 -22 -21 -20 -19 -18 -17 -16 -15

d13C (%0)

d15N

(%

)

Agriculture (<53) Fores t (<53) Agriculture (<53) Fores t (<53)

•A hotspot model for NRCS people

WEPP Model Calibration/verification

Watershed Soil Characterization

Property Units Corn Soybean CRP Floodplain Bank

Geological            

Silt % 65.4 59.2 63.3 70.0 66.4

Clay % 29.5 34.7 30.3 26.4 26.7

Sand % 5.10 6.10 6.40 3.60 6.90

Water Content % 21.5 20.0 25.36 16.1 18.35

Specific Gravity --- 2.56 2.73 2.46 2.54 2.50

Plastic Limit % 26.70 27.00 24.20 32.35 24.36

Liquid Limit % 36.34 38.07 38.59 47.00 37.68

Watershed Soil Characterization

Property Units Corn Soybean CRP Floodplain Bank

Chemical            

pH --- 7.70 7.75 6.05 6.45 6.95

Buffer pH --- 7.30 7.35 6.70 7.00 7.13

Exch. K cmol/kg 0.749 0.639 0.431 1.154 0.248

Exch. Ca cmol/kg 21.21 31.13 10.82 12.52 12.00

Exch. Mg cmol/kg 3.63 3.26 2.18 3.36 2.98

Exch. Na cmol/kg 0.07 0.10 0.05 0.03 0.04

Zn g/kg 0.0021 0.004 0.0011 0.0052 0.0016

Fe g/kg 0.070 0.098 0.116 0.140 0.088

Mn g/kg 0.013 0.010 0.018 0.021 0.017

Watershed Soil Characterization

Property Units Corn Soybean CRPFloodplai

n Bank

Chemical            

Organic Matter g/kg 43.55 54.85 53.85 74.70 30.52

Total C g/kg 23.85 30.05 29.59 40.96 16.71

Total N g/kg 2.061 1.964 2.672 3.496 1.638

NO3-N g/kg0.003

6 0.00220.002

6 0.0027 0.0038

NH4-N g/kg0.001

3 0.01400.004

0 0.0050 0.0080

CEC cmol/kg25.66

0 35.12017.08

9 17.069 15.266

SAR √(cmol/kg)

0.0191 0.0236

0.0205 0.0123 0.0135

Biological            

Photosynthetic Pathway --- C4 C3 C3  ---  ---

•30% clay, 65% silt and ~ 5% sand. If clay is 20-40%, no stable aggregates formed, thus higher erosion.•pH range=6.0-7.0, range in which particles are most susceptible to erosion due to F to F contacts between clay particles•Cation Exchange Capacity were higher in cultivated soils i.e., corn = 25.7 cmol/kg and soybean = 35.1 cmol/kg than uncultivated soils.

Watershed Soil Characterization

•Sodium adsorption ratio (SAR) was low for all soils which means higher erosion rate. •Mineralogy of clay – Majority Smectite, Vermiculite (and a small portion Kaolinite and Illite) have more water retention, hence lower shear required for soil erosion•Analyzed soil biogeochemical characteristics indicate that the catchment soil is highly erodible.

Watershed Soil Characterization

•Beginning with the 25 years simulation, equilibrium conditions for both water discharge and sediment discharge is reached

Results

•The estimated avg. annual water discharge at the outlet is 5,775,717 m3/yr, sediment discharge is 26,335 ton/yr, and sediment delivery ratio is 0.183

Results

Sample tray detail

Bank sediment placed in sample tray detail

Figure 6. A laboratory flume equipped with a sediment box sampler for testing critical erosional strength. Top to the right a bank sediment sample placed in the box sampler for testing its critical erosional strength.

DSD Generated by the Rainfall Simulator

0.01

0.1

1

10

100

1000

10000

0 2 4 6 8 10

Drop diameter (mm)

Dro

p S

ize

Dis

trib

uti

on

(m

m-1

m-3

)

RR = 80 mm/h

M-P DSD 80mm/h

M-P=Marshall Palmer DSD

ONE DIMENSIONAL HYDRODYNAMIC/SEDIMENT

TRANSPORT MODEL TO SIMULATE RILL EROSION

PROCESSES

Importance of Rills

Sketch of upland erosional processes

Geomorphologic Features

• Headcuts• Plunge pools• Steps: areas of locally flat and steep slopes• Rough, uneven bed profiles• Varying cross-sectional geometries

(constrictions and expansions) • The process of rill evolution involves a feedback

loop between erosion, hydraulics and bedform.

Geomorphologic Features

Example of a step pool sequence often observed in rills

Rill Erosion

• Rill erosion is a function of the capacity of the flow to detach sediment in relation to the capacity of the soil to resist detachment.

• Depth-averaged velocity, discharge, and bed shear stress are used to express the capacity of the flow.

• Rill sediment material is comprised of fines (clay, sand, silt, and organic material)

• This material can exist as either fine suspended sediment or as aggregates of many particles that move as bedload (Gilley et al., 1990).

Primary Objectives and Goals

The objective of this research is to develop a numerical model that removes some of the limitations of the existing models and accounts for the complex interaction of flow, geomorphology and sediment transport. The model proposed here is limited to 1-D flows and consists of a hydrodynamic and sediment transport component which are solved simultaneously.

Model Description: Hydrodynamic Component

• 3ST1D (Steep Stream Sediment Transport 1-D) was previously developed for steep gravel bed streams.

• It employs the TVD-MacCormack Scheme.

• The Scheme solves the unsteady form of the St.Venant equations.

1 – D St. Venant Equations

A = flow cross-sectional area

Q = flow rate

g = gravitational acceleration

I1 = Hydrostatic Pressure force Term

I2 = Forces exerted by channel wall contractions or expansions

S0 = Bed Slop

Sf = Friction slope

= Momentum Coefficient β

Conservative Form

The conservative form of the equation guarantees correct jump intensities and celerities of surface waves (Lax and Wendroff, 1960).

TVD-MacCormack Scheme

• Used to solve the 1-D St. Venant equations

• Shock Capturing

• Is a two step predictor-corrector scheme

• Second order accurate and capable of rendering the solution ocsillation free (MacCormack, 1969)

Predictor Step

Δt = time stepΔx = cell sizei = computational point j = time leveln = Manning’s Roughness R = hydraulic radius

Corrector Step

The corrector step uses the values of U and H from the predictor step

Updating Step

Sketch of the Domain of the Model

Results: Rill Bed (200 cells)

-0.1

-0.05

0

0.05

0.1

0.15

0.2

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

Longitudinal distance (m)

Ele

vatio

n (m

)

Fixed Bed Profile

Measured Water Surface

Predicted Water Surface

Locations of Cross-sections

1 2

76

54

3

8

Max error in water surface: Pools = 22% Step = 10%

Profiles

Streamwise Distribution of Velocities (200 cells)

Error in the predicted velocity values exceed 40% in the pools

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

Longitudinal Distance (m)

Vel

ocit

y (m

/s)

Averaged Measured Velocity Predicted Velocity

Froude Numbers (200 cells)

0

0.5

1

1.5

2

2.5

3

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

Longitudinal Distance (m)

Fro

ude

Num

ber

Measured Froude Number Predicted Froude Number

Flow Continuity (200 Cells)

0

0.0002

0.0004

0.0006

0.0008

0.001

0.0012

0.0014

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

Longitudinal Distance (m)

Dis

char

ge (

m^3

/s)

3ST1D Predicted Flow Exact Flow

Max Error in flow continuity 9%

Future Work

1) Incorporation of a channel dynamic model into WEPP for modeling flow and sediment in ditches and channels

2) Enhancement of the soil properties databases3) Tracers for sediment provenance4) There is still a need to define spatial variability of

infiltration and runoff along a hillslope using new research methodologies and techniques.

5) In the Midwest the role of fertilizers (e.g., ammonia, manure), pesticides and etc. needs to be identified

6) Roughness characterization

QUESTIONS ?