WORKPLACE DOCUMENT, UNDER PREPARATION: /data/work/helena/hohen/hohmsc1.html

http://www2.gis.uiuc.edu:2280/modviz/protected/hohen/hohmsc4.html

go back to main document hoh4.html

Multiscale, process based, distributed water and sediment transport simulation

Helena Mitasova, Lubos Mitas
University of Illinois at Urbana-Champaign

1. Introduction

With the growing capabilities to collect digital data about landscape from different sources using different technologies it is common that the data characterizing the studied landscape are very heterogeneous with different coverage, resolution, detail and accuracy. To fully exploit the data available for given area for modeling (predictions and simulations) algorithms should be able to handle and make effective use of such heterogeneous data sets, maximize the acuuracy and detail of predictions while performing computations efficiently.

2. Methods

Methods and algorithms used for multiscale simulation are described in the following papers

• Multiscale: L Mitas 1998 Vienna
• Mitas in preparation
• SIMWE: Mitas, L., Mitasova, H., 1998, Distributed erosion modeling for effective erosion prevention Water Resources Research Vol. 34, No. 3, pp. 505-516.

3. Results

For SIMWE application, which is computationally more demanding than USPED the study watershed is represented by 400x400 DEM at 10m resolution. Small subarea has available 1m contours interpolated to 2m DEM (area is XxX at 10m resolution and 680x395 at 2m). For comparison/evaluation we ran also simulations for the entire waterhsed at 2m resolution 2000x2000 DEM. Simulations including the 1997 cover data are further complicated by the fact that, the cover data are available at 20m resolution

Location of the subregion selected for a high resolution and multiscale study 1997

Simulations for uniform rainfall, soil and cover conditions, including multiscale:

• modeling and visualization of water flow over 3D terrain with spatially variable resolution
  
• topographic potential for erosion and deposition at 10m resolution(compare with the USPED 10m and 2m result)
  
• multiscale topographic potential for erosion and deposition at 10m/2m resolution
  

Simulations for spatially variable conditions in 1997

Input parameters:

• elevationa and its gradient (dx,dy)
  
• rainfall excess f(r-i(K)) 1997
  
• roughness 1997
  
• erodibility f(C*K) 1997
  
• transportability 1997
  

Results:

• for fully saturated soil (no infiltration)
  • sediment flow rate
    zoom into intensive use area
    q>10: high rates of sediment flow = 3.5% of total area
    
  • net erosion deposition 1997
    zoom into intensive use area
    high risk areas 5.9% of
    total area
    
• for spatially variable infiltration (high in undisturbed areas)
  • sediment flow rate
    zoom into intensive use area
    q>10: high rates of sediment flow = 3.5% of total area
    
  • net erosion deposition 1997
    zoom into intensive use area
    high risk areas 3.9% of
    total area