Process based distributed erosion modeling using SIMWE

Lubos MITAS, Helena MITASOVA
University of Illinois at Urbana-Champaign

DRAFT

1. Study area

http://www.edv.agrar.tu-muenchen.de/fam/hello_fam.html
http://www.edv.agrar.tu-muenchen.de/fam/scheyern/ressource.html

2. Methods

http://www.cecer.army.mil/grass/viz/SF.final/mitas.html
add here illustrated computation with maps of: elevation, direction of flow, slope, velocity (magn), water depth, water discharge shear stress, stream power, detachment capacity, transport capacity, .....

3. Parameters explained by spatially uniform case

(generalized 2D version of analysis presented by Wischmeier 1969 for profile) The base situation for all examples is 36mm/hour rainfall intesity, full saturation, rough surface (dense grass), sandy soil (negiligible critical shear stress, sigma=1, low detachability(erodibility) and transportability Kt=Kd=0.0003) Parameters are described in depth in the WEPP manual.

Rainfall excess (rainfall intensity - infiltration) influences the magnitude of erosion/deposition rates:

r=36mm/h r=72mm/h r=144mm/h

Surface roughness (represented by mannings n, chezy or darcy-weisbach) influences ration of spatial extent of erosion/deposition, the rougher the surface the greater the spatial extent of deposition

n=0.01 n=0.10

(keep Kt=Kd=konst, change only n)

Critical shear stress (or critical stream power) has impact on ersoion/deposition pattern, its high value reduces the spatial extent of erosion, however it can increase the magnitude of erosion rates in steeper and/or convergent areas. (light green areas have no detachment=no erosion)

tauc=1. tauc=3. tauc=5. tauc=7.

Erodibility (detachment capacity coefficient) Changes spatial distribution of erosion/deposition, does not change the magnitude of sediment load in stream

movie, Kd=(0.0001,0.1), Kt=0.001(oprav)

Transportability (transport capacity coefficient) Changes both spatial distribution and magnitude of sediment flow as well as erosion/deposition rates significantly.

movie, Kt=(0.0001,0.1), Kd=0.001(oprav)

The parameters do not act independently, they are interrelated and it is the interaction which controls the pattern and magnitude of erosion. For example of both Kd and Kt change at the same rate, only magnitude of erosion/deposition changes and the spatial distribution stays the same. Illustration of a realistic change for the uniform case (WRR paper)

4. Impact of spatial variability in land cover

4.1Impact of differences between the properties of two different covers and/or soils at different location within a complex terrain using theoretical combinations of parameters.
rainfall excess (due to difference in infiltration)
roughness 0.1/0.05, 0.1/0.01
erodibility (detachment coef. Kd) 0.03/0.0003, 0.003/0.0003
transportability (Kt) 0.03/0.0003, 0.003/0.0003
sigma Kd/Kt
tauc

4.2 Observed and simulated erosion depposition for original land use for:
a) small event and dense cover
b) large event and bare soil

Compare with colluvial deposits, linear erosion features, profiles(yields, Cs), density of rills-erosion potential

4.3Simulated water depth, sediment flow rate and net erosion/deposition for the orginal, new(sustainable) and computer designed land use, for the time of the year when the agricutural land (brown) is bare and grass cover(green) is well developed ( Table 1. Values of parameters used in this simulation).

  Land cover        Water            Sediment flow    Erosion/deposition

A : traditional land use (21% grass)
B : new sustainable land use (41% grass)
C : computer generated land use (19% grass)

Note: rerun water for B to reach steady state

Note the high erosion around buffers where the difference in Mannings is high ( add a different view)

Both B and C hold more water, reduce the max. sediment flow rates and max erosion rates. Note the proper simulation of deposition in the upper parts of grass areas, and increased erosion below the grass contour strips: phenomena which are difficult to model by more traditional methods.

5. Impact of temporal variability in land cover

1. n=0.01/0.1, Kt=Kd=0.03/0.0003
2. n=0.05/0.1, Kt=Kd=0.03/0.0003
Note: when n changes, also Kt needs to be modified(?) to reflect change in transportability for rougher surface? Explore changes only n - surface is rougher but soil is as transportable as before. Change both n and kt - increased cover cause increase in roughness and decrease in transportability(?)

this allows us to evaluate the effectiveness of land use design under different cover conditions, as the vegetation develops over the year.

References

This work was partially supported by DOD Strategic Environmental Research and Development Program. ( SERDP).

GMSL Home Page

GMSL Modeling & Visualization Home Page

Contacts

Helena Mitasova (GMSLab)
helena@gis.uiuc.edu Bill Brown (GMSLab)
brown@gis.uiuc.edu Lubos Mitas (NCSA)
lmitas@ncsa.uiuc.edu