SPATIAL MODELS OF SOIL PROPERTIES

HELENA MITASOVA, LUBOS MITAS, WILLIAM M. BROWN
GMSLab, University of Illinois at Urbana-Champaign

DRAFT !

1. Introduction:

problem (Mitas and Mitasova 1997) soil analysis is needed for assesing the fertilizer need, suitability for certain crop or vegetation, assessing polution and erosion damage (inputs for erosion model, impact of erosion on soils) etc.

2. Methods

Interpolation: 2D and 3D RST (regularized spline with tension and smoothing: theory and properties; see Mitas and Mitasova 1988, Mitasova and Mitas 1993, Mitasova et al. 1995, Mitas and Mitasova 1997, Mitas and Mitasova: in preparation) link to interp.html

Visualization:

• Improved query algorithm.
  Developed code for improved 3D surface query performance. Query function allows use of arbitrary clipping planes (in addition to the parallelepiped view volume) and multiple surfaces at variable resolutions, each with a unique mask.

  The basic algorithm is as follows:

  1. transform selected point on view plane to a view ray
  2. intersect view ray with convex polyhedron defined by the intersection of the parallelepiped view region with any user defined clipping planes.
  3. if ray enters this polyhedron, trace ray to find any intersections with visible (unmasked) parts of any surfaces
     Note: View ray is traced to find intersection with surface by following the projection, or "shadow" of the ray along each polygon of the surface until the shadow passes above the view ray (when viewing from above the surface). The previous implementation traced the view ray itself, until a point below the surface was found and then used binary recursion to pinpoint the intersection. Sharp peaks and spikes were easily missed and the selection of the inital "step" down the view ray was critical and needed to be adjusted for various data ranges and exaggerations. The new strategy results in faster point selection and an exact intersection.
  4. choose closest intersection to viewer (or return an ordered list)
  5. query the database (data sources for color and terrain) directly using geographic coordinate of intersection(s).

  Such point-on-surface functionality is useful for 3D data querying, setting center of view, placing geographic objects or scales, and setting center of rotation for vector transformations.

• Relative scaling.
  Multiple surfaces are quite useful to visualize boundaries of layers. For example, surfaces may be created that represent soil horizons so that thickness of layers may be displayed. This presents a technical challenge in terms of dimensional scale. To study differences between two similar surfaces, we use a scaled difference approach where only the spatial distance between surfaces is exaggerated, maintaining correct surface intersections. In other words, one or more surfaces are scaled relative to a base surface. This concept of a base surface (e.g., bedrock or surface terrain in soil studies) has been integrated within visualization tools to make it easier to experiment with selection of the base surface. Similarly, 3D volume data from orthogonal regions may be warped to lie along a (possibly exaggerated) base surface, to explore the effect of terrain on the volumetric feature being modeled.
• Volume visualization.
  Volume visualization has been implemented on top of the new GRASS "grid3" data access library, resulting in the two programs r3.mkdspf and r3.showdspf. These programs will be combined and incorporated into the primary visualization tool, NVIZ, which has been ported to use the OpenGL graphics library standard. Use of the grid3 library should allow us to resample volume data on the fly, enabling more interactive volume exploration. The following volume visualization methods will be incorporated:
  1. isosurfaces - currently precomputed by r3.mkdspf and displayed by r3.showdspf, will be computed as needed at selectable resolution and for smaller subregions and displayed within NVIZ.
  2. fence diagrams - currently available for multiple surfaces within NVIZ and for volume data using r3.showdspf, will be integrated within NVIZ.
  3. volume rendering - will be implemented using 3D texture extension to OpenGL and optimized volume slicing.

3. Results

All volume models have vertical exageration 100

Soil horizons

sites movie land use

Long term traditional land use had a significant impact on spatial distribution of chemicals in soil, as shown in the following examples.

Soil chemistry analysis:

a) soil reaction (ph)

• multiple surface representation of ph draped over the terrain surface and individual horizons

  hor 1 hor 3 hor 5

  and by blended cutting planes

  y=100     y=140     

• volume representation

  isosurfaces

  movie acid neutral

  planes

  slicing fence diagram (row=40,80,120,160)

• 3D interpolation for 2D horizon (test)

Volume model incorporates the vertical relationship into interpolation and allows more efficient visual analysis

The highest acidity is on terrain surface in grass area and it extends over most of the area in deeper horizons

b) organic carbon

• surface representation: % organic carbon draped over terrain

                         

• volume model

  isosurfaces box-movie

The highest concentration of organic matter is in the long term grass area. The amount of organic matter rapidly decreases with depth.

c) plant available K

• surface representation (draped on terrain)

• volume model

  isosurfaces fence diagram

Area has surplus of Potassium with maximum concentrations in grassy area (serves as filter/accumulation area?). Concentrations decrease with depth, except for a hop field in valley where the higher concentrations extend well bellow the A horizon.

d) plant available P

• surface representation (draped on terrain)

• volume model

  isosurfaces fence diagram

Area has surplus of Phosphates, with maximum concentrations in valleys (depressions) Concentrations decrease with depth...? (volume model, solid: opt, surplus)

Location of optimal and surplus K, P : intersection of solids

• optimal Kc,Pc surplus Kc,Pc

Note that the optimal concentrations od K,P cover only a small surface area (find better visualization for this) and they are located mostly bellow surface in lower horizons.

e) total nitrogen

• surface representation (draped on terrain)

• volume model

  isosurfaces fence diagram

f) bulk density

• surface representation (draped on terrain)

• volume model

  isosurfaces movie

Size fraction analysis:

• point symbol model for all fractions

  
  fractions at sampling sites

• surface model

  sand [%] clay [%]

• volume model (sand, clay)

  clay [%] clay [%]

  high% low[%]

  Movie through all fractions???

Soil color:

Soil texture and structure:

qualitative/descriptive data, use point symbols

Derived soil parameters:

a) hydraulic conductivity

The values of hydraulic conductivity were derived for each 3D point based on the particle size distribution using equations from WEPP manual. The values were then interpolated to a 3D raster which can be used as an input for 3D infiltration model or visualized using isosurfaces or crossections:

• surface representation                   

• volume representation

4. Application to land management

5. Conclusion

Highest concentrations, acidity, chemicals in grassy area - does it serve as depository/filter? K,N,P different behavior as they have different location of max. concentration....

6. References

The document is copyrighted, send requests to use the images to helena@gis.uiuc.edu

GMSL Modeling & Visualization
Home Page