EROS

1
EROS (Crave Davy, 2001)
Stochastic model of erosionsedimentation
processes, based on cellular automata,
which mimics the natural variability of climatic
events with deterministic transport processes

• Very simple concept model based on water flux
  conservation and mass balance
• Not rigid properties such as channel geometry
  emerge from the action of stochastic processes
• Produces realistic discharge patterns and
  fluvial network morphologies

• - Landscape evolution is driven primarily by
  sediment transport ? transport-limited
• - It is somehow difficult to relate the model
  parameters to measurable quantities

2
CHILD (Tucker at al., 2001)

• User friendly
• Includes a wide range of fluvial incision laws
  and tectonic scenarios
• Uses a Triangulated Irregular Network
• Stochastic rainfall variability

• - No landsliding algorithm
• - Some properties are relatively rigid, e.g.
  channel geometry defined by hydraulic scaling
  relationships

3
CHILD (Tucker at al., 2001)
Uplift rate increased 1 Ma ago
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
4
CHILD (Tucker at al., 2001)
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
5
Profile shape and landscape morphology
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Fiamignano Fault
Time 0.0 My
Questions can we reproduce the catchments
morphology using a simple detachment-limited
fluvial erosion law (model testing)? What is the
effect of dynamic channel adjustment
(sensitivity analysis)?
Fault
Uplift
6
Profile shape and landscape morphology
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 0.2 My
7
Profile shape and landscape morphology
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 0.4 My
8
Profile shape and landscape morphology
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 0.6 My
9
Profile shape and landscape morphology
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 0.8 My
10
Profile shape and landscape morphology
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 1.0 My
11
Profile shape and landscape morphology
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 1.2 My
12
Profile shape and landscape morphology
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 1.4 My
13
Profile shape and landscape morphology
Basin internally drained
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 1.6 My
Slope lt 0
14
Profile shape and landscape morphology
Basin internally drained
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 1.8 My
Slope lt 0
15
Profile shape and landscape morphology
Basin internally drained
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 2.0 My
Slope lt 0
16
Profile shape and landscape morphology
Basin internally drained
Example 1 landscape response to tectonic
perturbation application to the Central
Apennines, Italy (Attal et al., 2008)
Time 2.2 My

• Simple DL model makes relatively good
  predictions.
• Role of CHANNEL NARROWING (morphology response
  time)

Slope lt 0
17
CHILD (Tucker at al., 2001)
No vegetation
Example 2 (sensitivity analysis) effect of
vegetation and wildfires on landscape
development application to the Oregon Coast
Range (Istanbulluoglu Bras, 2005)
? Vegetation cover and changes in cover through
time (landslides, wild fires) strongly affects
landscape morphology (relief, drainage density)
Static vegetation cover
(Roering et al., 1999)
18
Modelling landscape evolution
Where are we? Models are getting more and more
complex, include more and more processes and
parameters, but Some essential key issues need
to be addressed ? Role of sediment (fluvial
erosion), ? Role of life (vegetation,
bioturbation, etc.), ? Role of climate
19
Climate and landscape evolution
? Climate is highly variable at geological time
scales
? Climate (e.g. temperature, rainfall,
storminess) strongly influences erosion rates and
processes
Hillslopes landsliding, freeze-thaw cycles
Rivers erosion occurs during discrete events
floods which mobilize sediments
20
Climate and landscape evolution
Problem most studies involving landscape
evolution modelling assume that climate
parameters are constant over millions of years!
Solution coupling landscape evolution models
with climate models
Problem SCALE!!!
LANDSCAPE EVO. MODELS
TIME STEP 10s hours to 1000s year
GRID RESOLUTION 10s to 1000s meters
CLIMATE MODELS
TIME STEP 15 minutes to a few hours
MIN GRID RESOLUTION 100 km!
21
Vision for the future?
The aim

• Solving the scale problems to couple climate and
  landscape development models
• realistic predictions of landscape evolution as
  a result of climate variability feedbacks
  between topography and climate
• predictions at varied time and space scales
  (from minutes to millions of years, from a small
  stream to whole countries!)

http//www.cru.uea.ac.uk/cru/info/ Climatic
Research Unit, UEA Norwich
22
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
PAUSE
Climate parameters
Hillslope transport landslide threshold
Initial topography
Fluvial sediment transport deposition bedrock
erosion
Additional parameters and algorithms fluvial
hydraulic geometry, bedrock and sediment
characteristics, role of vegetation, etc.
CHILD
23
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
24
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
25
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
26
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
27
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
28
(No Transcript)
29
(No Transcript)
30
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
31
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
If OPTDETACHLIM 1 ? E KAmSn
32
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
To set in the .in file
t ktqmbSnb kt(Q/W)mbSnb
Calculated by CHILD
We will consider Erosion ? Specific Stream power
(Law 2) E ? O/ W ? E K Q S / W ? mb 0.6
nb 0.7 pb 1.5 tc 0
To set in the .in file
E kb (t tc)pb
kt 1197 (typical value for bedrock rivers) kb
or kr poorly constrained parameters
33
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
34
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
35
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
CHILD user guide
36
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
Channel Width is calculated at each time step for
every node
W kwQ?sQb(?b-?s) kwQ?s if ?s ?b
37
The Channel-Hillslope Integrated Landscape
Development (CHILD) model (Tucker et al., 2001)
38
Visualizing the output using Matlab
ctrisurf(filename, ts, 0)
39
(No Transcript)