Hydrologic modeling of reconstructed watersheds using a system dynamics approach
Jutla, Antarpreet Singh
The mining of oil sands in the sub-humid region of Northern Alberta, Canada causes large-scale landscape disturbance, which subsequently requires extensive reclamation to re-establish the surface and subsurface hydrology. The reconstructed watersheds examined in this study are located at the Syncrude Canada Limited mine site, 40 km North of Fort McMurray, Alberta, Canada. The three experimental reconstructed watersheds, with nominal soil thicknesses of 1.0 m, 0.50 m and 0.35 m comprised a thin layer of peat (15-20 cm) over varying thicknesses of secondary (till) soil, have been constructed to cover saline sodic overburden and to provide sufficient moisture storage for vegetation while minimizing surface runoff and deep percolation to the underlying shale overburden. In order to replicate the hydrological behavior, assess the sustainability, and trace the evolution over time of the reclaimed watersheds, a suitable modeling tool is needed. In this research, a model is developed using the system dynamics approach to simulate the hydrological processes in the three experimental reconstructed watersheds and to assess their ability to provide the various watershed functions. The model simulates the vertical and lateral water movement, surface runoff and evapotranspiration within each watershed. Actual evapotranspiration, which plays an important role in the hydrology of the Canadian semi-arid regions, is simulated using an indexed soil moisture method. The movement of water within the various soil layers of the cover is based on parametric relationships in conjunction with conceptual infiltration models. The feedback relationships among the various dynamic hydrologic processes in the watershed are captured in the developed System Dynamic Watershed Model (SDWM). Most hydrological models are evaluated using runoff as the determining criterion for model calibration and validation, while accounting for the movement of moisture in the soil as a water loss. Since one of the primary objectives of a reconstructed watershed is to maintain the natural flora and fauna, it is important to recognize that soil moisture plays an important role in assessing the performance of the reconstructed watersheds. In turn, soil moisture becomes an influential factor for quantifying the health of the reconstructed watershed. The developed model has been calibrated and validated with data for two years (2001-2002), upholding the sensitive relationship between soil moisture and runoff. Accurate calibration of the model based on simulations of soil moisture in the various soil layers improves its overall performance. The model was subsequently used to simulate the three sub-watersheds for five years, with changing the calibrated model parameters to use them as indicators of watershed evolution. The simulated results were compared with the observed values. The results of the study illustrate that all three watersheds are still evolving. Failure to identify a unique parameter set for simulating the watershed response supports the hypothesis of watershed evolution. Soil moisture exchange between the till and peat layers changed with time in all of the watersheds. There was also a modest change in the water movement from the till to shale layers in each of the sub-watersheds. Vegetation is increasing in all of watersheds although there is an indication that one of the sub-watersheds may be sustaining deep rooted vegetation. The results demonstrate the successful application of the system dynamics approach and the developed model in simulating the hydrology of reconstructed watersheds and the potential for using this approach in assessing complex hydrologic systems.
DegreeMaster of Science (M.Sc.)
SupervisorKells, James A.; Elshorbagy, Amin A.
CommitteePeng, Jian; Maule, Charles P.; Barbour, S. Lee; Pomeroy, John W.
Copyright DateJanuary 2006