Soil Geomorphology and Landscape Modeling in South-Central Minnesota

Location

Continent: North America
Country: United States of America
State: Minnesota
City/Town: Waseca
UTM coordinates and datum: none

Setting

Climate Setting: Polar
Tectonic setting: none
Type: Process

Click the images for a full-sized view.

Show Figure 1 caption

Hide

Block diagram of the Clarion-Nicollet-Webster catena showing the distribution of different soil types on the landscape. USDA NRCS Soil Survey.

Show Figure 2 caption

Hide

Photographs of representative soil profiles of the Clarion-Nicollet-Webster catena. Notice the gradient of soil color from oxidizing conditions at the summit (red/brown subsurface colors) to reducing conditions at the toeslope (gray subsurface colors). Images courtesy of Jay Bell and the University of Minnesota Soil Landscape and Analysis Lab.

Show Figure 3 caption

Hide

Example of a high resolution digital elevation model (DEM). Lighter colors represent areas of lower elevation, while darker colors represent areas of higher elevation. The large area of low elevation in the upper center is a quarry. Image courtesy of the University of Minnesota Soil Landscape and Analysis Lab.

Show Figure 4 caption

Hide

Percent slope derived from the DEM in Fig. 4. Lighter colors represent areas of low slope, while darker colors represent ares of high slope. Notice the areas of high slope near the quarry. Image courtesy of the University of Minnesota Soil Landscape and Analysis Lab.

Show Figure 5 caption

Hide

Mean values for each terrain attribute of the Clarion-Nicollet-Webster catena. Collectively these data indicate that each soil type has a unique landscape signature. Modified from Swanson (2003).

Description

The unique and distinctive nature of soil reflects the cumulative interaction of the climate in which it is developing, the parent materials it is forming from, biological processes, the duration of soil development, and the configuration of the landscape. On a local scale, where most factors are relatively constant (climate, parent material, biologic processes and time), much of the variation in soil characteristics is a function of the nature of the landscape. The discipline of soil geomorphology deals with characterizing the relationships between soils and the landscape. "Soils and landforms develop together, and soil geomorphology is designed to examine and elucidate the nature of that genetic "dance". This development is a two-way street. Soils are affected by landforms, and through their developmental accessions and features, they in turn influence geomorphic evolution" (Schaetzl and Anderson, 2005).

A catena is a sequence of soils along a hillslope from the summit to the toeslope. The variation in soil characteristics observed in a catena is mainly due to topography and its influence on sediment and water movement. In particular, differences in hydrology along a hillslope leads to predictable and generally visually obvious gradients of soil color. Where water tables are deep and oxidizing conditions are present (summit and shoulder positions), subsurface soil colors are red or brown from ferric (oxidized) iron. Conversely, in areas where water tables are close to the surface (footslope and toeslope positions) soil colors are gray, reflecting lower irons content due to translocation of ferrous (reduced) iron under reducing conditions. In addition, soils in reducing environments often have a thick black surface horizon due to slow decomposition of organic materials. Along a gradient of soil hydrology, a catena of unique soil types can be defined. An example of a representative soil catena from south-central Minnesota can be seen in Figs. 1 and 2. Notice the gradient of soil colors from the summit (oxidized) to the toeslope (reduced). This sequence of soils formed on calcareous loamy glacial till deposited on a low-relief ground moraine during the Wisconsinan glaciation (~15,000 ybp).

Until recently, communication of soil-geomorphic relationships has been largely accomplished through conceptual and qualitative models, such as Fig. 1. With the emergence and advancement of geographic information systems and spatial datasets, we now have the ability to develop quantitative models. Characterizing the geomorphic surface is the first step in producing a quantitative soil-landscape model. A Digital Elevation Models (DEM) is a digital representation of land surface topography (x, y, and elevation). An example of a high-resolution DEM is shown in Fig. 3. Using a DEM, "terrain analysis" can be used to derive quantitative "terrain attributes" that describe the nature of the land surface. These variables include: plan and profile curvature, slope, aspect, flow length, flow accumulation, catchment area, and others. Figure 4 shows an example of percent slope derived from the DEM in Fig. 3. These terrain attributes provide a powerful way to visualize topography. It is well established that soil patterns reflect the nature of the land surface (i.e. soil catena). By identifying the soil types and comparing them to the terrain attributes, we can develop a quantitative model of soil distribution on the landscape. Ideally each soil type will have a unique terrain signature. In addition, soil-landscape models can also be used to model any soil property that varies spatially in response to topography, such as soil thickness, thickness of the A-horizon (topsoil), soil water content, depth to carbonates (extent of leaching), organic carbon content, pH, and surface texture.

A soil-landscape model was developed for the Clarion-Nicollet-Webster soil catena shown in Figs. 1 and 2 (Swanson, 2003). Terrain attributes were derived from a 5-m resolution DEM (Fig. 4). Each soil type had a distinct landscape signature and within each terrain attribute the values varied along a gradient from the summit (Clarion) to the toeslope (Glencoe) (Table 1). For instance, Glencoe, a very poorly drained soil, had profile and curvature values indicating concavity and flow accumulation, which fits our conceptual understanding of where we would expect to find reducing environments. Overall, the model that was developed was able to predict soil distribution with a classification accuracy up to 74%. Of the misclassified points more than 95%, were classified in a neighboring soil type along the catena. This proved to be a highly accurate model for predicting soil distribution in this region.

Associated References:

• Schaetzl, R. and S. Anderson. 2005. Soils: genesis and geomorphology. Cambridge University Press, New York. 817 p.
• Swanson, H. 2003. Using quantitative landscape modeling and scientific visualization to characterize soil hydrology dynamics. M.S. Thesis, University of Minnesota, St. Paul, Minnesota.

Supporting URLs