The goal of this thesis was to develop a better understanding of the drivers of stream equilibrium state in urban settings. Eleven reaches in ten Central Ohio watersheds were surveyed and analyzed to fulfill two specific research objectives: 1.) Create a geomorphology based dynamic equilibrium evaluation index that could determine the equilibrium state of a stream without the use of the bankfull stage and 2.) Determine the key landscape factors that cause a stream to be in or out of equilibrium.
Chapter 2 includes the methods and results for research objective one. Geomorphology data from each site was compared using the 1-year and 2-year recurrence interval discharges as well as the top of the bank. A qualitative analysis of the streams led to an initial classification of the study sites into two categories, in equilibrium or out of equilibrium. Binary logistic regression was then used to determine the best model for equilibrium classification. No single model was found to be better than all others. Eight geomorphology variables were found to be significant on their own and two, top of bank to 1-year ratios of discharge and unit stream power, were able to classify 10 of the 11 reaches correctly. Nearly all 2-parameter models were able to predict equilibrium with 100% success. These results show the feasibility of this classification approach, however, tests on a large independent dataset would be necessary to confirm the validity of the models and to choose a single best model.
The landscape study to fulfill objective two is discussed in Chapter 3. Three types of landscape factors, urbanization, attachment, and area for adjustment, were explored to determine which indicators would most affect the equilibrium state of the receiving stream. Binary logistic regression was used to compare the landscape data to the qualitative classification from Chapter 2 in order to choose the best equilibrium state predictors. A secondary analysis of a new reach on Rush Run was performed to better explain the effects of local landscape variables.
Results showed that watersheds with large amounts of recent or total urban area would result in streams that were out of equilibrium but that positive landscape features such as a wide adjustment area and a well-connected floodplain could mitigate the urban effects. Total developed area in a watershed was found to be the most important landscape variable as it was able to correctly explain the equilibrium state of all sites. All adjustment and attachment variables also predicted all sites correctly when combined with total developed area.
These results can help to explain some of the effects of urbanization on urban stream systems and serve as a guide for future land management decisions. With lower overall watershed development, good attachment, and a wide buffer zone a stream will be much more likely to maintain dynamic equilibrium despite urban development in its containing watershed.