Date of Award
2009
Level of Access Assigned by Author
Campus-Only Thesis
Degree Name
Master of Science (MS)
Department
Earth Sciences
Advisor
Andrew S. Reeve
Second Committee Member
Paul H. Glaser
Third Committee Member
Willem Brutsaert
Abstract
The potential significance and susceptibility of peatlands in future climate change has urged a further understanding of the hydrologic processes that determine their development and distribution. Hydrology, a primary control that influences and is influenced by carbon-based gas sequestration in peatlands, will only be understood when the hydraulic conductivity (K) is properly estimated. The K of the peat in these important carbon reservoirs affects flow patterns and rates that determine their spatial distribution and influence biogenic gas production and release from these ecosystems. Biogenic gasses, such as methane, within the peat complicate this relationship by lowering K of the peat and occluding pore space. Another factor inhibiting proper understanding of peatland hydrology is the scale at which variability in K occurs. Traditional methods of determining the spatial and directional hydraulic properties of peat yield data that are representative of small volumes and do not reflect the small-scale variations in K that influence water and solute transport in the subsurface. To investigate the small-scale variations of K in peat, a three-dimensional model was constructed using FiPy, a finite-volume modeling library under development at the National Institute of Standards and Technology (NIST). To characterize the distribution of K, a methodology of spatial parameter definition and estimation based on pilot points and accompanied by the use of regularization was used to solve the inverse problem. Automated parameter estimation was undertaken with PEST, a program that facilitates the calibration process, using data collected during pumping tests conducted in the Glacial Lake Agassiz Peatland (GLAP) in northern Minnesota. One test was conducted on a raised bog and the other on a Sphagnum lawn. Dependency of K with depth was found at both sites; however the relationship is not monotonic as assumed in several studies. Modeled K values between 7.71xl0"5 and 4.65xl06m/s were found at the bog site with localized areas of lower K material. Modeled K values between 2.66xl0"3 and 3.52xl0"5 m/s were found at the lawn site with the presence of lower K material located over higher K layers. While the K values determined through modeling are commensurate with some previous studies that have used other methods of measuring K, their three-dimensional distribution suggest a potential importance in a further understanding of peatland hydrology. The K values obtained from this study for deeper peat indicate its significance in the transmission of water for regional models. Furthermore, the differences in K between the sites reflect the hydraulic distinction between the different types of peat landforms. Lastly, the presence of low K zones found at both sites provide a unique look into the subsurface hydraulic structure and may provide further evidence to the development and existence of mobile free-phase gas within peat.
Recommended Citation
Rhoades, Joshua L., "Characterizing the Distribution of Hydraulic Properties in the Glacial Lake Agassiz Peatlands Using a Three-Dimensional Numerical Model and Regularized Inversion" (2009). Electronic Theses and Dissertations. 775.
https://digitalcommons.library.umaine.edu/etd/775
Files over 10MB may be slow to open. For best results, right-click and select "save as..."