Assessing positional accuracy and correcting point data for digital soil mapping at varying scales

Minerva J Dorantes, Purdue University

Abstract

Accuracy, timeliness, and the effect of scale of soil maps are rarely assessed. The recent increase in the use of GIS technologies and modelling software in natural resources and land management, has increased the demand for soil information at a finer resolution worldwide. Most of the world's developing countries rely on soils information at a scale that is too coarse for practical planning, and have obstacles impeding collection of new data, such as civil war and a lack of collection resources. The United States has an exhaustive collection of soils data at a fine scale. However, its location information is replete with errors and inconsistencies which, if unaccounted for, can affect predictive model estimates. An integrated digital soil mapping methodology is necessary to extract the wealth of knowledge stored in soil survey data for building detailed soil maps and for assessing the positional accuracy of soil pedon data. Two studies were conducted using public data contained in the U.S. Soil Survey databases. The first study tested the development of an accurate regional-scale digital soil class map by combining new elevation data and satellite imagery. As a result, a model design was created that may be applied in countries with limited soil data. In the second study, several models were developed to assess the locational accuracy of the U.S. Soil Survey pedon points for Indiana. The study resulted in the creation of a more detailed Public Land Survey System grid, as well as several ArcGIS tools to assign a margin of error to existing soil pedon point locations, which separately or together can be adopted on a national scale.

Degree

M.S.

Advisors

Owens, Purdue University.

Subject Area

Agronomy|Physical geography|Soil sciences

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS