Date of Award:

5-2014

Document Type:

Dissertation

Degree Name:

Doctor of Philosophy (PhD)

Department:

Plants, Soils, and Climate

Committee Chair(s)

Janis L. Boettinger

Committee

Janis L. Boettinger

Committee

Astrid R. Jacobson

Committee

Lawrence E. Hipps

Committee

Thomas C. Edwards Jr.

Committee

Jürgen Symanzik

Abstract

Soil information is required for arid and semi-arid land management decisions such as permitting livestock grazing or planning vegetation restoration projects. However, traditional soil mapping methods may not provide adequate soil information, because the scale of mapping often requires dissimilar soils to be grouped together and there are no estimates of map uncertainty. Traditional methods are also often too costly or impractical to implement in large, remote, public arid and semi-arid rangelands. Digital soil mapping (DSM) may be able to overcome these limitations. Digital soil mapping is the creation of pixel-based soil maps using quantitative statistical models that relate easily measured biophysical environmental variables derived from geospatial data (e.g., slope and aspect from a digital elevation model) with more difficult to measure soil observations.

We investigated DSM for producing soil information useful for land management decisions. Specifically we: 1) compared multiple DSM methods for predicting soil taxonomic classes, 2) predicted the spatial distribution of potential biological soil crust classes, and 3) measured threshold friction velocity, a necessary input for wind erosion modeling.

Many existing soil use and management decisions are based on soil taxonomic classes; thus digital soil taxonomic class maps are useful for quantitative decision making. However, there are a large number of available DSM methods to produce such maps. Understanding which DSM method produces the most accurate soil taxonomic maps would contribute to robust management decisions. Comparison of DSM methods revealed that prediction accuracy was more dependent upon the number of taxonomic classes and the number of observations of each taxonomic class, than the specific method chosen.

Biological soil crusts (BSC) are important organisms in arid lands, but are highly susceptible to surface disturbance. Maps of BSC potential (BSC in the absence of disturbance) would be useful for understanding the impact of different land uses on BSC distribution. Digital soil mapping can be used to make such maps. We produced maps of low, moderate, and high potential BSC level-of-development classes. Accuracy assessment revealed that only the moderate level-of-development class was reliable. The map of the moderate BSC level-of-development class is anticipated to be useful for assessing the impact of land use practices on BSC distribution.

Proposed pumping in western Utah could reduce groundwater, thus reducing vegetation cover and exposing more soil surface area to wind erosion. Evaluating the potential impacts of proposed pumping requires the use of wind erosion models. Such models require inputs of threshold friction velocity (TFV), which is the minimum turbulence required to initiate wind erosion. However, TFV is difficult to measure, and we sought to predict TFV from easier to measure soil surface properties. We found TFV to be dependent upon soil surface type. Only undisturbed soils with weak physical crusts and some undisturbed soils with surficial rock fragments reached TFV. All soil surfaces reached TFV when disturbed. On average, soils with weak physical crusts were more susceptible to wind erosion, but great variability between surface types was found. Threshold friction velocity in undisturbed soils with weak physical crusts and undisturbed soils with surficial rock fragments was predicted using a combination of penetrometer force, percent rock cover, and silt concentration.

Checksum

3ed330e52a3e92c3d1936579e894515a

Download

Share

COinS

Copyright for this work is retained by the student. If you have any questions regarding the inclusion of this work in the Digital Commons, please email us at .