Peatland data analysis and simulation with Python
Authors: Reeve, Andrew (University of Maine); Westervelt, Claire (University of Maine)
Over the past five years, data has been collected to assess the hydrology and carbon dynamics of the Red Lake Peatlands in northern Minnesota. Python and associated software (eg. cython, pytables, scipy) have been an integral tool for analyzing this multi-year data set containing over 30 million values, and for the creation of computer models used to simulate peat deformation, groundwater flow, and peat accumulation. Data collected from the Red Lake Peatlands include GPS measurement of surface movement, hydraulic head measurements from monitoring wells, and a range of weather parameters. Python tools have facilitated and automated data management and analysis, allowing the identification of 1) hydrology-driven surface deformation and 2) anomalous data suggesting rapid release of free-phase biogenic gas. In addition, two computer models have been created using python tools to test ideas and assess fundamental processes within peatland systems. The first model assumes the peat column behaves like a Kelvin-Voigt viscoelastic material and is used to assess surface movement. The second model couples peat accumulation to groundwater flow to evaluate the long term development of peatland systems.