"Uncertainty is inevitable when creating any kind of model. A model can be used in the most accurate way possible, if the uncertainties are understood. This study determines the level of uncertainty in the Bank Stability and Toe Erosion Model (BSTEM) of the Osage River downstream of Bagnell Dam between Lake Ozark, MO and Jefferson City, MO. The statistical analysis of the BSTEM model was performed using the aid of SAS statistical computer modeling software. There were 4 different analysis values used to determine the best fit model for all dependent variables. These values include the F-test, the coefficient of determination, mean squared error, and Mallow's CP. The F-test is used to determine that there is indeed a relationship between the independent variables and the dependent variables, whereas the other 3 values help narrow down the simplified statistical models to determine the best fit model for each dependent variable. There were 4 different BSTEM outputs that were used in the uncertainty analysis. These 4 dependent variables are average applied boundary shear stress, factor of safety, maximum lateral retreat and eroded area - total. The statistical analysis determined how many best fit statistical models each variable appeared in and this information helped to determine the variables affecting the BSTEM model. The variables that appeared in all the best fit statistical models had a large impact on the BSTEM model, whereas the ones that did not show up in a best fit statistical model had a small effect on the BSTEM model. The factor of safety analysis yielded results that were inconclusive, while the other three variables had a confidence level ranging from 76.7% up to 90.6%, with an average confidence of over 80% for the entire BSTEM model"--Abstract, page iii.
Morris, Charles Darwin
Cawlfield, Jeffrey D.
Civil, Architectural and Environmental Engineering
M.S. in Civil Engineering
Missouri University of Science and Technology
vii, 60 pages
Osage River (Mo.)
© 2013 Annabell Leigh Ulary, All rights reserved.
Thesis - Open Access
Library of Congress Subject Headings
Riparian areas -- Missouri -- Osage River
Hydraulic models -- Case studies
Erosion -- Missouri -- Osage River -- Mathematical models
Electronic OCLC #
Ulary, Annabell Leigh, "Analysis of model uncertainty in hydraulic modeling: the BSTEM application to the Osage River" (2013). Masters Theses. 5366.